Skip to main content

Full text of "The modern locomotive"

See other formats 


UC-NRLF 


THE 

MODERN 
LOCOMOTIVE 


C  EDGAR  ALLEN 


The  Cambridge  Manuals  of  Science  and 
Literature 


THE    MODERN    LOCOMOTIVE 


CAMBRIDGE   UNIVERSITY   PRESS 

2LonU0tt:   FETTER  LANE,   E.G. 

C.  F.  CLAY,  MANAGEB 


100,  PRINCES  STREET 

ILontion:   WILLIAM  WESLEY  AND  SON,  28,  ESSEX  STREET,  STRAND 

Berlin:   A.  ASHER  AND  CO. 

SUtpjtg:   P.  A.  BROCKHAUS 

$eto  lorft:   G.  P.  PUTNAM'S  SONS 

Botnbag  ant)  Calcutta :  MACMILLAN  AND  CO.,  LTD. 


All  rights  reserved 


THE    MODERN 
LOCOMOTIVE 


C.  EDGAR  ALLEN 


A.M.I.Mecfa.E. :   A.M.I.E.E. 


Cambridge : 
at  the  University  Press 

New  York: 
C.  P.  Pbtnajn's  Sons1 


1912       ;%  ,, 


(Eambrtoge: 


PRINTED    BY    JOHN    CLAY,    M.A. 
AT    THE    UNIVERSITY    PRESS 


With  the  exception  of  the  coat  of  arms  at 
the  foot,  the  design  on  the  title  page  is  a 
reproduction  of  one  used  by  the  earliest  known 
Cambridge  printer,  John  Siberch,  1521 


Jt  r* 


PREFACE 

TN  a  small  book,  not  intended  for  specialists  but 
for  a  wider  public,  it  has  not  been  possible  to  do 
more  than  sketch  the  general  principles  governing 
the  design  and  working  of  a  modern  locomotive,  and 
to  trace  the  broad  lines  of  development  from  its  com- 
paratively simple  predecessor  of  twenty-five  or  thirty 
years  ago.  More  attention  has  been  given  to  such 
matters  as  combustion,  transfer  of  heat,  steam  produc- 
tion, superheating,  compounding,  feed- water  heating, 
resistance  and  stability,  as  being  essential  to  the 
proper  understanding  of  the  modern  locomotive,  than 
to  mechanical  features,  to  do  justice  to  which  would 
involve  a  mass  of  technical  detail.  That  phase  of  the 
subject  which  catalogues  the  dimensions  of  various 
types  of  engines  actually  in  service  has  also  been 
avoided. 

249447 


viii  PREFACE 

The  author  appreciates  his  indebtedness  to  many 
of  the  books  and  proceedings  mentioned  in  the  biblio- 
graphy, and  he  desires  to  acknowledge  the  information 
supplied  by  Mr  J.  G.  Bo  wen  Cooke,  Chief  Mechanical 
Engineer  of  the  L.  &  N.  W.  Railway,  and  Mr 
W.  P.  Reid,  Chief  Mechanical  Engineer  of  the  North 
British  Railway.  His  thanks  are  due  to  the  Editor 
of  the  Engineer  for  permission  to  reproduce  the 
illustration  on  page  24.  In  a  similar  manner  he  is 
indebted  to  Messrs  Constable  &  Co.,  Messrs  Doin  et 
Fils,  Paris,  and  to  that  valuable  work,  The  Loco- 
motive of  To-day,  for  the  illustrations  on  pages  55, 
109  and  43  respectiv-ely. 

C.  E.  A. 


WATFORD, 

December,  1911. 


CONTENTS 

CHAP.  PAGE 

Introduction 1 

I.  Steam  generation.     The  boiler    .        .        .  12 

II.  Combustion  and  vaporization      ...  30 

III.  Increasing  the  useful  effect  of  the  boiler  .  41 

IV.  Superheating,  thermal  storage,  feed  heating  63 

V.  Resistance,  tractive  effort,  adhesion   .        .  79 

VI.  Utilization  of  the  steam      ....  93 

VII.  Frames  and  running  gear    .        .        .        .  112 

VIII.  Stability 125 

IX.  Performance  and  speeds      .        .        .        .  141 

X.  Compounding 157 

Bibliography 170 

Index     ........  172 


THE  MODERN  LOCOMOTIVE 


INTRODUCTION 

FEW  subjects  possess  more  importance  for  the 
public  to-day  than  travelling,  and  it  is  safe  to  say 
that  in  the  mind  of  the  majority  this  is  primarily 
associated  with  the  railway  and  more  vaguely  with 
the  locomotive.  Long  familiarity  with  the  locomotive, 
coupled  with  a  general  sameness  in  its  external  ap- 
pearance, has  engendered  indifference  and  rendered 
all  who  have  not  given  special  attention  to  the  subject 
oblivious  to  the  fact  that  many  and  radical  changes 
have  been  taking  place  in  its  design.  Moreover 
its  hold  on  public  interest  has  to  some  extent  been 
challenged  by  a  younger  rival,  the  electric  locomotive 
or  electrically  propelled  train,  whose  possibilities  are 
generally  credited  with  an  exaggerated  importance. 
Many  no  longer  express  astonishment  at  the  tale  of 
the  achievement  of  the  '  iron  horse/  but  are  inclined 
to  call  into  question  its  easy  pre-eminence  for  hauling 
fast  and  heavy  trains.  If  put  to  it  however,  they 

A.  L.  1 


S  THE  MODERN  LOCOMOTIVE 

will  confess  to  a  sentimental  interest  aroused  by  the 
attributes  of  strength,  symmetry  and  self-sufficiency 
possessed  by  the  locomotive,  and  would  confess  to  a 
pang  of  regret  if  it  were  ultimately  displaced  in 
the  fierce  conflict  of  less  noble,  if  not  less  efficient, 
competitive  systems. 

In  spite  of  all  that  is  heard  to  the  contrary,  one 
feature  stands  out  very  clearly  in  the  mind  of  those 
qualified  to  form  an  accurate  judgment,  that  is,  there 
is  no  reasonable  prospect  of  our  trusted  friend  being 
relegated  to  the  scrap  heap  or  becoming  an  isolated 
relic  allowed  to  stand  in  solitary  grandeur  on  a 
concrete  foundation  in  leading  railway  stations.  The 
limitations  of  electric  traction  are  far  too  clearly 
recognized  and  easily  defined,  and  unless  some  un- 
foreseen and  revolutionary  change  takes  place  the 
steam  locomotive  will  be  found  doing  its  duty  many 
years  hence.  Here  it  may  be  stated  that  the  true 
object  of  the  electrification  of  railways  is  the  diversion 
of  passengers  from  competing  tramways  or  omnibuses, 
or  the  development  of  populous  districts,  both  of 
which  are  local  problems  and  have  little  in  common 
with  the  problem  of  the  steam  working  of  main-line 
traffic. 

For  the  reason  of  the  existence  of  the  locomotive 
we  must  look  back  far  into  the  seventeenth  century, 
when,  in  the  remote  colliery  districts  of  the  North, 
coal  haulage  was  laboriously  effected  by  wagons 


INTRODUCTION  3 

slowly  dragged  along  wooden  tramroads  by  horses. 
The  feeble  resistance  of  wooden  rails  to  wear  and 
their  susceptibility  to  decay  led  to  their  displacement 
by  rails  made  of  iron,  which  were  widely  adopted 
in  most  of  the  colliery  districts. 

While  horse-power  or  the  stationary  steam  engine 
remained  the  only  tractive  force  available  for  the 
haulage  of  wagons,  a  fixed  limit  was  placed  on  the 
development  of  the  railway  system.  It  was  the 
invention  of  the  locomotive  by  Trevithick  and  its 
subsequent  improvement  by  the  Stephensons,  that 
was  to  give  a  great  impetus  to  the  construction  of 
iron  roads,  and  place  the  question  of  steam  locomo- 
tion on  a  successful  basis.  It  is  only  when  we  hold 
steadily  in  view  the  position  occupied  by  railways 
to-day  as  compared  with  their  humble  origin  of  a 
century  ago,  that  we  are  enabled  to  realise  how 
much  we  owe  to  the  locomotive  and  how  greatly 
economic  history  is  bound  up  with  it. 

The  writer's  intention  is  not  to  cover  the  ground 
of  more  or  less  ancient  history  and  examine  in  de- 
tail the  gradual  development  of  the  locomotive  to 
its  present  dimensions  and  design ;  it  will  suffice  to 
say  that  the  original  engines  of  the  Stockton  and 
Darlington  or  the  Liverpool  and  Manchester  Rail- 
ways would  be  regarded  as  mere  toys  compared 
with  a  modern  express  engine.  Their  relative  size 
and  capacity  are  well  illustrated  by  the  photograph 

1—2 


4  THE  MODERN  LOCOMOTIVE 

reproduced  in  the  frontispiece  to  this  volume  shewing 
the  Rocket,  or  rather  an  exact  reproduction  of  that 
famous  engine,  alongside  one  of  Mr  Bowen  Cooke's 
latest  express  engines  on  the  London  and  North 
Western  Railway. 

It  was  in  the  middle  fifties  that  design  settled 
down  to  the  definite  type  with  the  four  coupled 
driving  wheels  and  a  leading  pair  of  carrying 
wheels,  which  was  to  remain  for  so  long  the  standard 
practice  of  Great  Britain  and  the  Continent.  This 
wheel  arrangement,  with  the  exception  that  a  four- 
wheel  bogie  has  replaced  the  pair  of  leading  wheels, 
may  be  said  to  predominate  even  to-day.  From  the 
period  mentioned  until  1 870  nothing  very  remarkable 
in  the  way  of  new  development  took  place,  but  the 
ensuing  twenty-five  years  witnessed  the  introduction 
of  many  fresh  and  notable  locomotive  types.  It  is 
a  tribute  to  the  excellence  of  those  engines  that  a 
number  of  them — witness  the  Precedent  class  of  the 
London  and  North  Western  Railway,  rebuilt  it  is  true 
— are  still  running.  The  Charles  Dickens  (No.  955), 
the  most  famous  of  this  class,  left  the  Crewe  shops 
in  1882  and  ran  daily  from  Manchester  to  Euston 
and  back,  a  distance  of  366  J  miles,  completing  her 
millionth  mile  9  years  219  days  after.  During 
that  period  she  made  2651  runs  between  the  towns 
mentioned  in  addition  to  92  other  journeys,  and 
consumed  12,515  tons  of  coal. 


INTRODUCTION  5 

Charles  Dickens  continued  to  work  the  8.30  a.m. 
Manchester  to  London  and  4.0  p.m.  London  to  Man- 
chester trains  until  August  5th,  1902,  on  which  date 
the  engine  completed  2,000,000  miles  on  the  return 
journey  from  London  to  Manchester,  having  run 
5312  trips  to  London  and  back,  in  addition  to  186 
other  sundry  trips.  The  engine  was  withdrawn  from 
the  Manchester  to  Euston  express  passenger  service 
on  August  5th,  1902  :  it  has  since  been  working 
passenger  trains  between  Manchester,  Crewe,  Shrews- 
bury, Birmingham  and  Leeds,  and  has  occasionally 
run  goods  trains.  The  total  number  of  miles  run 
by  this  engine  from  the  date  of  first  turning  out  of 
the  works  in  February,  1882,  up  to  and  including 
October  31st,  1911,  was  2,318,918. 

Other  notable  engines  of  this  period  were 
Mr  D.  Drummond's  four-coupled  engines  on  the 
Caledonian  ;  Mr  W.  Stroudley's  Gladstones  on  the 
London, Brighton  and  South  Coast ;  Mr  S.  W.Johnson's 
single  wheelers  on  the  Midland,  and  Mr  Patrick 
Stirling's  8  ft.  singles  on  the  Great  Northern. 

But  if  the  period  above  mentioned  witnessed  the 
production  of  a  number  and  variety  of  new  types 
what  must  be  said  of  tne  last  twelve  years  ?  The 
speed  of  trains  has  materially  increased,  the  demand 
for  the  most  rapid  transport  of  both  passenger  and 
goods  has  become  more  urgent,  and,  further,  the 
taste  of  the  public  coupled  with  competition  amongst 


6  THE  MODERN  LOCOMOTIVE 

the  railway  companies  have  been  responsible  for  the 
introduction  of  coaches  of  larger  dimensions — some 
upwards  of  35  tons  in  weight :  all  of  these  develop- 
ments have  enormously  added  to  the  demands  made 
upon  the  locomotive.  A  Great  Western  express  fifty 
years  ago  weighed  no  more  than  60  tons  behind  the 
tender,  whereas  a  modern  express  weighs  anything 
from  200  to  350  tons,  to  say  nothing  of  the  tender 
itself  which,  when  no  water  pick-up  apparatus  is 
used,  weighs  from  45  tons  upwards  instead  of 
15  to  25  tons.  The  following  table  and  Fig.  3  shew 
the  gradual  increase  of  locomotive  dimensions. 

But  designers  in  their  endeavours  to  solve  the 
problem  were  face  to  face  with  a  grave  difficulty, 
namely,  the  limited  dimensions  imposed  by  the 
strictly  limited  loading  and  structure  gauge  within 
which  any  improvement  could  be  made. 

In  this  respect  British  engineers  have  been  at 
a  great  disadvantage  as  compared  with  some  of  their 
foreign  and  American  rivals,  and  many  of  them  have 
more  than  once  regretted  that  the  '  battle  of  the 
gauges'  was  to  the  4  ft.  8J  in.  party,  and  have,  in  spite 
of  its  many  attendant  drawbacks,  sighed  for  the  old 
broad  gauge  dimensions  which  would  have  permitted 
largely  increased  boiler  and  cylinder  dimensions  and 
more  room  for  the  'motion'  and  lengthened  bearings. 
Or,  to  state  the  case  in  other  words,  with  the  stan- 
dard gauge  the  difficulties  are  accentuated  by  the 


INTRODUCTION  7 

fact  that  the  space  allowed  outside  the  rails  is  less 
in  proportion  than  for  narrow-gauge  engines. 


Fig.  1.     The  Jenny  Lind. 


Fig.  2.     A  Precedent  engine.    London  and  North  Western  Railway. 


THE  MODERN  LOCOMOTIVE 


II 


Tt<          CO          O 

t-         CO         <N 

I          -^         to         »O 


s 


I         I       9      ¥>      ? 

I  I          IO         t>-         !>•         to 

r-t          rH          iH          <M 


S    g    2 


o      o      o      o      o 


CO         i-l         CO          CO 

t~      co      co      co 


•^         O        "^         CO        "^         CO        CO  ^•v 
(M<M(M(N(M<M<N't 

XXXXXXX§ 

O         »O         CO         GO         C~         O5         "5  pH 


MI 


...  1^9 

TfS'iSS  tnPn 


INTRODUCTION 


Fig.  3.     The  Great  Bear ;  a  Stirling  Single,  and  the  Locomotion 
shewn  to  the  same  scale. 

Another  feature  of  recent  years  has  been  the 
demand  for  long  distance  non-stop  trains  travelling 
over  one  hundred  miles  without  a  stop,  as  for 
example  Euston-Crewe,  Paddington-Plymouth,  St 
Pancras-Shipley  :  this  has  also  taxed  the  ingenuity 
of  locomotive  engineers. 


10  THE  MODERN  LOCOMOTIVE 

To  meet  this  combination  of  demands,  the  past 
decade  has  seen  a  surprisingly  large  number  and 
variety  of  new  types  culminating  in  such  giants  as 
the  White  Bear  on  the  Great  Western  and  the 
Baltic  class  on  the  Northern  of  France. 

At  the  same  time  this  production  has  been 
accompanied  by  a  steady  tendency  towards  specializa- 
tion of  duty,  the  conditions  gradually  bringing  the 
locomotives  of  different  companies  more  and  more 
in  accord  where  the  work  to  be  done  is  similar. 

Coincident  with  this  feature  of  similarity  in 
arrangement,  we  notice  that  the  heating  surfaces  of 
boilers  have  been  extended,  grate  areas  have  been 
proportionately  enlarged  and  steam  pressures  have 
been  raised  by  as  much  as  fifty  per  cent.  One  point 
worthy  of  comment,  however,  is  that  practically  no 
increase  of  cylinder  dimensions  has  been  made. 
Locomotive  engineers  are  by  no  means  agreed  that 
increased  cylinder  dimensions  are  desirable  and,  as 
explained  above,  there  are  practical  objections  to 
the  construction  of  locomotives  with  cylinders  of 
exceptional  diameter ;  hence  endeavours  in  the 
direction  of  greater  capacity  have  been  confined  to 
the  multiplication  of  the  cylinders  themselves.  Four- 
cylinder  locomotives  have  been  tried  on  the  London 
and  South  Western  and  Great  Western,  and  the 
Lancashire  and  Yorkshire  railways,  and  the  latest 
express  engine  on  the  North  Eastern  Railway  has 


INTRODUCTION  11 

three,  an  arrangement  which  has  also  been  adopted  on 
the  Midland  and  Great  Central.  The  multiplication 
of  cylinders  is  not  here  referred  to  in  connection  with 
the  compound  system  in  which  four  cylinders  exist 
in  a  very  large  number  of  locomotives  in  Europe  and 
America.  Usually  there  are  two  inside  cylinders 
and  two  outside ;  the  low-pressure  cylinders  are 
generally  outside  and  the  high-pressure  cylinders 
inside  the  frames.  An  interesting  exception  is  pro- 
vided by  the  Italian  State  Railway,  in  which  the  two 
high-pressure  cylinders  are  both  on  the  left-hand  side 
and  the  two  low-pressure  cylinders  both  on  the  right. 
A  general  examination  of  the  highly  interesting  com- 
pound system,  which  has  been  looked  to  by  many  as 
a  means  of  securing  a  higher  efficiency  is  however 
reserved  for  treatment  in  a  separate  chapter. 

In  the  modern  engine  the  number  of  driving 
axles  is  usually  two  or  three  and  a  leading  bogie 
is  the  general  rule.  The  Great  Western  Railway  has 
however  a  large  number  with  a  leading  carrying  axle. 

Bavaria  first  and  now  France  have  adopted  a 
leading  and  trailing  bogie  on  the  same  engine. 

The  gross  weight  and  adhesive  weights  have 
reached  high  figures,  amounting  in  the  first  case  to 
ninety  tons,  and  in  the  second,  with  locomotives 
with  two  driving  axles,  forty,  and  where  three  are 
used,  forty-five  tons.  In  exceptional  cases  weights 
exceeding  eighteen  tons  per  axle  are  found. 


12  THE  MODERN  LOCOMOTIVE         [CH. 

Reference  may  be  made  to  the  devices  used  in 
the  Continental  engines  for  reducing  air  resistance, 
in  which  the  front  of  the  smoke  box  and  the  cab 
front  are  made  wedge-shaped ;  also  to  the  means 
recently  adopted  to  increase  the  useful  effect  of  the 
engine,  such  as  superheating,  feed-water  heating 
and  thermal  storage,  which  however  receive  special 
consideration  in  a  later  chapter. 

In  the  result  British  locomotive  engineers  have 
dealt  with  the  problem  presented  to  them  with 
conspicuous  success,  and  it  is  abundantly  evident 
that  from  the  splendid  performances  of  which  the 
details  are  constantly  being  made  public  in  railway 
literature,  that  high  rates  of  speed  with  maximum 
loads  are  attainable  whatever  be  the  form  of  engine 
adopted,  provided  that  due  regard  is  paid  to  the 
essentials  of  scientific  design  and  construction. 


CHAPTER  I 

STEAM  GENERATION.    THE  BOILER 

IT  is  not  too  much  to  say  that  the  success  of 
an  engine  depends  entirely  on  the  boiler,  since  the 
power  developed  is  limited  by  the  amount  of  steam 
it  is  capable  of  supplying.  The  constantly  increasing 


i]  STEAM  GENERATION  13 

demands  for  more  powerful  engines  have  led  to  a 
corresponding  development  of  boiler  dimensions  to 
the  extent  that  the  limits  imposed  by  the  loading 
gauge  have  almost  been  reached.  Important  re- 
strictions hamper  the  locomotive  designer  in  that 
the  length  of  boiler  is  governed  almost  entirely  by 
the  wheel  base,  and  the  diameter  by  its  relation  to 
that  of  the  driving  wheels  ;  further  a  large  quantity 
of  high-pressure  steam  is  necessary,  which  means 
that  the  volume  of  water  to  be  evaporated  in  a 
given  time  is  also  considerable.  Limitations  of  grate 
area  involve  intense  combustion  induced  by  forced 
draught,  and  a  large  heating  surface  must  be  pro- 
vided to  utilize  satisfactorily  the  heat  thus  generated. 
Thus  the  problem  to  be  faced  is  a  more  difficult  one 
than  that  occurring  in  stationary  or  marine  practice. 
Generally  speaking,  modern  practice  has  not 
meant  any  great  departure  from  the  form  of  boiler 
possessed  by  the  earliest  locomotives,  and  to-day 
we  have  for  essentials  the  features  possessed  by 
Stephenson's  Rocket,  which  so  effectually  disposed  of 
its  rivals  in  the  Rainhill  trials  by  reason  of  its  quick 
steam  production.  For  its  success  it  depended  upon 
the  adoption  of  the  multitubular  principle,  the  idea 
of  which  originated,  not  with  the  Stephensons,  but 
with  Mr  Booth,  secretary  of  the  Liverpool  and  Man- 
chester Railway,  and  with  Seguin  in  France.  This 
use  of  a  large  number  of  tubes — usually  from  200 


14  THE  MODERN  LOCOMOTIVE          [OH. 

to  250 — forms  the  essential  feature  of  the  modern 
boiler  and  is  the  chief  means  of  enabling  the 
conditions  mentioned  above  to  be  carried  out. 

Four  elements  compose  the  locomotive  boiler ; 
the  inner  and  outer  fire-boxes,  smoke-box,  and  a 
cylindrical  body  containing  the  tubes,  which  latter 
run  from  the  inner  fire-box  to  the  smoke-box.  The 
heated  gases  from  the  fire  travel  along  the  tubes  to 
the  smoke-box  and  communicate  heat  to  the  water 
which  surrounds  both  the  inner  fire-box  and  tubes. 
It  will  thus  be  seen  that  a  large  heating  surface  is 
obtained  by  employing  a  considerable  number  of 
tubes. 

Referring  to  Fig.  4,  A  is  the  inner  fire-box.  It  is 
roughly  rectangular  in  shape,  the  sides  and  crown 
are  rolled  from  one  sheet  of  metal,  usually  about 
T9e  in.  thick  throughout.  This  is  riveted  to  the  front 
tube-plate  B  and  the  back  plate  <?,  which  are  both 
flanged  on  three  sides  for  this  purpose.  The  metal 
generally  employed  is  copper,  because  it  does  not 
readily  oxidise  and  resists  the  action  of  the  fire 
better  than  steel.  The  latter  metal  is  now  much 
used  because  it  is  less  expensive,  and  has  the 
same  coefficient  of  expansion  as  the  material  of 
which  the  rest  of  the  boiler  is  composed.  The 
bottom  of  the  inner  fire-box  is  formed  by  the 
grate,  consisting  of  a  number  of  wrought  iron  or 
steel  firebars,  K,  through  the  spaces  between  which 


STEAM  GENERATION 


15 


the  air  necessary  for  combustion  finds  its  way  to  the 
fire.  The  firebars  are  usually  fin.  thick  on  their 
upper  surface  and  from  4  to  4J  ins.  deep,  tapering 


Fig.  4.    Details  of  a  modern  fire-box. 


16  THE  MODERN  LOCOMOTIVE         [OH. 

off  to  |  in.  at  the  bottom.  Thickening  pieces  at 
their  ends  and  centres  keep  them  the  necessary 
distance  apart.  Under  the  fire-box  is  the  ashpan,  L, 
provided  with  dampers,  M,  in  front  and  behind,  for 
regulating  the  admission  of  air.  Ashpans  are  of 
ample  dimensions  to  prevent  accumulated  ashes  from 
interfering  with  the  air-supply,  and  generally  the 
bottom  is  made  to  contain  water  for  quenching  the 
ashes.  Control  of  the  damper  doors  is  by  means  of  a 
handle,  <7,  fixed  on  the  fireman's  side  of  the  footplate. 

Inside  the  fire-box  above  the  grate  and  just 
below  the  bottom  tubes  is  a  firebrick  arch,  H, 
which,  becoming  intensely  hot,  assists  combustion 
and  directs  the  hot  gases,  so  that  they  impinge  on 
the  sides  and  crown  of  the  fire-box.  In  this  it  is 
assisted  by  the  deflection  plate,  G,  fitted  opposite 
above  the  fire-hole  or  on  the  fire-door. 

The  tube-plate  B  is  drilled  to  receive  the  tubes. 
For  this  reason  it  is  made  of  thicker  plate  than  the 
crown  and  sides,  i.e.  up  to  1  in.,  or,  with  steel  as 
the  material,  it  would  be  about  half  this  thickness. 
To  increase  its  resistance  to  the  action  of  the  fire 
an  alloy  of  nickel  and  copper  has  been  tried,  also 
a  combination  of  copper  for  the  lower  portion  and 
steel  for  the  upper  part  which  receives  the  tubes. 

The  outer  fire-box  or  shell,  N9  is  composed 
usually  of  three  steel  plates  from  J  in.  to  ^  in. 
thick,  a  wrapper  forming  the  sides  and  top,  the 


i]  STEAM  GENERATION  17 

throat  plate  in  front  which  receives  the  boiler  barrel, 
and  the  back  plate  P  containing  the  fire-door.  The 
outer  and  inner  fire-boxes  are  strongly  connected 
at  the  bottom  by  a  foundation  ring  8,  the  ring  round 
the  fire-hole  T,  and  a  large  number  of  staybolts  Z>. 
The  latter  are  very  highly  stressed,  owing  to  the 
enormous  pressure  acting  on  the  inner  and  outer 
boxes  tending  to  thrust  them  apart.  The  magnitude 
of  this  pressure  will  be  realised  when  it  is  stated  that 
it  ranges  from  200  Ibs.  per  square  inch  upwards,  which 
means,  in  a  fire-box  of  average  dimensions,  a  total 
pressure  of  over  250  tons  on  the  crown  and  about 
400  tons  on  each  side.  Fractures  of  the  stays  are 
frequent  enough  and  are  due  principally  to  the 
bending  action  set  up  by  the  unequal  expansion  of 
the  inner  and  outer  fire-boxes.  A  double  influence 
produces  this  effect,  (1)  the  metal  of  the  inner  fire- 
box as  we  have  seen  has  a  coefficient  of  expansion 
different  from  that  of  the  outer ;  (2)  the  inner  fire- 
box reaches  a  higher  temperature. 

The  question  as  to  the  best  material  to  use  for 
staybolts  is  one  of  the  problems  of  the  day.  Copper 
is  in  general  use,  wrought  iron  and  steel  less  so  ;  and 
recently  phosphor  bronze  and  a  manganese-copper 
alloy  have  been  tried  with  good  results.  The  stays 
are  usually  from  J  to  1  in.  in  diameter  and  pitched 
4  ins.  apart,  so  that  each  supports  a  plate  area  of 
16  sq.  ins.  at  each  end. 

A.  L.  2 


18  THE  MODERN  LOCOMOTIVE         [OH. 

Staying  the  crown  of  the  fire-box  is  of  importance. 
Two  methods  are  in  use,  one  employing  direct 
stays  E  and  sling  stays  F  (Fig.  4),  and  the  other 
girder  stays.  The  latter  is  the  more  usual.  Each  has 
its  adherents,  those  favouring  girder  stays  claiming 
that  their  use  permits  a  greater  amount  of  freedom 
for  expansion  than  the  first  mentioned.  One  design 
of  girder  stay  is  shewn  in  Fig.  5.  The  girders  are 
of  /  section  and  support  the  roof,  their  ends  taking 
a  bearing  on  the  edges  of  the  inner  fire-box  as 
shewn,  the  stresses  being  transmitted  by  the  vertical 
plates  to  the  foundation  ring.  The  roof  bars  and 
fire-box  crown  are  connected  by  slings.  These  tend 
to  slacken  when  the  box  expands  on  first  heating, 
but  as  the  pressure  rises  they  are  put  in  tension. 

The  barrel  or  cylindrical  shell,  although  carrying 
an  enormous  pressure,  does  not  present  the  same 
difficulty  in  arranging  for  the  resistance  to  stress  as 
do  the  flat  surfaces  of  the  fire-box.  Stresses  in  a 
cylindrical  shell  are  easily  calculated,  as  is  the 
strength  of  riveted  joints,  and  it  is  thus  simply  a 
question  of  providing  plates  of  suitable  tensile 
strength  and  joints  equally  strong  against  failure. 

The  barrel  is  made  up  of  two  or  more,  often 
three,  rings  rolled  from  suitable  plates  about  T%  in. 
thick,  and  either  riveted  together  by  straps  or 
hoops,  or  telescoped,  each  ring  being  pushed  into 
its  neighbour,  as  shewn  in  Fig.  5.  The  latter  is  a 


I] 


STEAM  GENERATION 


19 


^.2.2  a 

02  02  02  02  02 


O2  OQ  O2  >• 


-    .    si 

<"     *     'fi'l, 

I 


fl      .  t>D  - 

.  a  <D  a     a*  .     OD 

^§  si      e' 


P-i  02 

pqpqoapq  SP^ 
2—2 


20  THE  MODERN  LOCOMOTIVE          [CH. 

favoured  method,  since  the  position  of  the  largest  ring, 
next  the  fire-box,  allows  a  liberal  water  space  and 
gives  a  better  provision  for  circulation,  both  of  which 
are  rendered  necessary  by  the  higher  temperatures 
incidental  to  the  higher  pressures  now  employed.  In 
addition  to  the  staying  power  afforded  by  the  tubes, 
longitudinal  stays  of  1 J  in.  diameter,  iron  or  steel, 
run  from  the  smoke-box  tube  plate,  which  is  from 
|  to  |  in.  thick,  to  the  back  plate  of  the  fire-box. 

The  junction  of  the  barrel  and  the  fire-box  is  one 
of  the  weakest  parts  of  the  boiler.  Several  methods 
are  in  vogue  to  secure  safe  connection  ;  these  how- 
ever need  not  detain  us. 

The  products  of  combustion  as  we  have  seen  are 
conducted  to  the  smoke-box  and  chimney  by  the  fire 
or  flue  tubes,  which  represent  the  largest  portion  of 
the  heating  surface  and  absorb  a  large  quantity  of 
heat ;  in  fact  the  difference  in  temperature  at  the 
back  and  front  ends  of  the  tubes  reaches  from  600 
to  700°  F.  The  tubes  generally  employed  in  this 
country  are  of  copper,  with  smooth  interior  and 
exterior  surfaces.  They  vary  in  outside  diameter 
from  If  in.  up  to  2  ins.,  and  are  usually  13  Standard 
Wire  Gauge  thick  throughout.  In  Europe  and 
America,  however,  iron  and  more  recently  soft  steel 
tubes  have  been  introduced.  Copper  has  the  ad- 
vantage over  steel  in  its  more  effective  resistance 
to  corrosion  due  to  hard  and  bad-quality  water ; 


i]  STEAM  GENERATION  21 

on  the  other  hand  copper  tubes  are  more  expensive. 
French  engineers  maintain  that  with  water  of  good 
quality  steel  tubes  are  preferable,  but  admit  that 
in  other  circumstances  the  question  is  doubtful.  In 
some  cases  steel  tubes  reinforced  with  copper  "  safe 
ends  "  at  the  fire-box  tube  plate  are  used. 

To  increase  the  available  heating  surface,  tubes 
known  as  Serve  tubes,  which  are  2  ins.  or  more  in 
diameter,  have  been  employed  to  some  extent  in 
recent  years. 


Fig.  6.     Cross  section  through  a  Serve  tube. 

A  section  through  such  a  tube  is  shewn  in  Fig.  6. 
It  will  be  seen  that  they  have  longitudinal  internal 
ribs  or  fins,  which  add  some  90  per  cent,  to  the 
internal  heating  surface.  Their  superiority  from  the 
point  of  view  of  vaporization  has,  however,  been 
contested.  The  ribs  also  obstruct  the  free  passage  of 
soot  and  cinders,  and  by  reason  of  their  rigidity  such 
tubes  set  up  severe  stresses  in  the  tube  plates.  The 
tubes  are  slightly  inclined  from  front  to  back  and  are 
arranged  in  vertical  columns,  which  facilitates  the 
dislodgment  of  the  steam  from  their  surfaces. 


22  THE  MODERN  LOCOMOTIVE         [CH. 

The  method  of  setting  tubes  in  the  tube  plate  is 
of  the  greatest  importance,  as  defective  work  and 
vibration  lead  to  leakage.  This  is  also  caused  sooner 
or  later  by  the  wearing  of  the  tube  ends  by  the 
abrasive  action  of  cinders,  deposits  of  scale,  and 
temperature  variations.  The  joint  (Fig.  7)  is  made 
by  expanding  the  tube  in  the  hole,  beading  over  the 
end  and  then  driving  in  a  ferule  of  Swedish  high- 
carbon  steel.  On  the  Northern  of  France  Railway, 





Fig.  7.     Details  of  a  joint  between         Fig.  8.     Method  of  jointing  a 
tube  plate  and  tube.  tube  employed  on  the  North- 

ern of  France  Kailway. 

which  uses  steel  tubes,  the  tubes  are  swaged  to  a 
smaller  diameter,  rolled  and  beaded  into  the  copper 
tube  plate  (Fig.  8). 

The  tubes  pass  into  the  back  plate  of  the  smoke- 
box,  which  forms  the  l front  end'  of  the  boiler. 
The  front  plate  is  fitted  with  a  large  dished  circular 
door,  accurately  fitted  to  close  up  air-tight.  It  is 
hinged  at  the  side  and  fastened  with  a  central  handle, 
which  actuates  a  number  of  dogs  pitched  equally 
round  the  periphery  of  the  door  and  locked  by  a 
second  handle  on  the  same  spindle.  The  door  permits 


STEAM  GENERATION 


23 


access  to  the  tubes  so  that  they  may  be  swept  or 
'run,'  it  also  enables  the  accumulated  ashes  in  the 
smoke-box  to  be  removed.  Formerly  and  even 
to-day  on  the  London  and  South  Western  and  other 
railways,  the  smoke-box  was  quite  short  (Fig.  9), 


Fig.  9.    Short  smoke-box,  and  spark  arrester  ;  Caledonian  Kailway. 

but  more  recently,  following  American  practice,  it 
has  been  much  lengthened  to  form  a  cylindrical 
projection  from  the  boiler  known  as  the  '  extended ' 
smoke-box  (Fig.  10).  The  means  for  obtaining 


24  THE  MODERN  LOCOMOTIVE         [CH. 

forced  draught,  namely  the  blast  pipe  P,  is  situated 
herein,  as  is  also  a  part  of  the  superheating  apparatus, 
S,  where  such  is  employed ;  further  the  steam  pipe  E 


Fig.  10.    Extended  smoke-box,  variable  blast  orifice  and  piston  valve. 
London  and  North  Western  Kailway. 


i]  STEAM  GENERATION  25 

from  the  boiler  to  the  steam  chest  passes  through  it, 
itself  acting  to  some  extent  as  a  superheater. 

The  primary  function  of  the  smoke-box  and  its 
equipment  is  a  most  important  one,  namely  the 
production  of  a  partial  vacuum  and  hence  a  draught, 
upon  which  depends  the  economical  burning  of  the 
fuel  and  at  such  a  rate  of  combustion  as  is  necessary 
for  satisfactory  steaming.  These  qualifications  depend 
largely  upon  the  disposition  of  the  blast-pipe  orifice 
in  its  relation  vertically  to  the  chimney  top,  together 
with  a  correct  height  from  the  boiler  centre  line  and 
a  correctly  proportioned  chimney. 

The  action  of  the  blast  pipe  as  at  present  under- 
stood is  as  follows  :  the  exhaust  steam  escaping  from 
the  cylinders  issues  through  the  contracted  orifice  of 
the  blast-pipe  at  a  high  velocity  and  expels  part  of  the 
contents  of  the  smoke-box,  thereby  creating  a  partial 
vacuum.  To  destroy  this  vacuum  the  products  of 
combustion  are  forced  through  the  tubes  by  the 
atmospheric  pressure  outside  the  ashpan  and  fire- 
hole,  thereby  creating  a  fierce  draught  on  the  fire. 
The  amount  of  draught  is  measured  by  the  vacuum 
in  the  smoke-box,  which  may  be  anything  up  to 
7  or  8  inches  of  water. 

Another  and  important  function  of  the  smoke- 
box  and  its  equipment  is  to  deal  with  the  enormous 
and  variable  quantity  of  air  needed  for  combustion 
which  is  drawn  in  at  the  fire-box.  This  air,  it  must 


26  THE  MODERN  LOCOMOTIVE          [OH. 

be  remembered,  is  immediately  expanded  six  to  eight 
times  by  rise  of  temperature,  and  on  arrival  at  the 
smoke-box  occupies  from  two  to  three  times  its 
original  volume,  the  difference  being  accounted  for 
by  the  rapid  cooling  which  takes  place  in  its  passage 
through  the  tubes. 

Mr  Hughes,  chief  mechanical  engineer  of  the 
Lancashire  and  Yorkshire  Railway,  carried  out  a 
year  or  two  back  some  interesting  experiments  with 
a  view  of  ascertaining  the  value  of  long  and  short 
smoke-boxes.  A  higher  vacuum  was  obtained  in 
every  case  with  the  extended  smoke-box  which,  as 
Mr  Hughes  points  out,  tends  to  prove  that  the  long 
box  serves  as  a  reservoir,  thus  assisting  the  mainte- 
nance of  draught  between  each  exhaust,  and  so 
modifying  the  intermittent  character  of  the  blast. 
This  was  also  verified  by  the  action  in  the  U-shaped 
glass  tubes  or  manometers  partially  filled  with 
coloured  water  used  to  observe  the  vacuum.  With 
the  extended  smoke-box  the  water  remained  quite 
steady,  and  only  moved  when  the  steam-discharge 
up  the  chimney  was  altered  ;  whereas  with  the  short 
box  the  water  was  in  a  constant  state  of  agitation, 
rising  and  falling  with  each  exhaust.  Further,  the 
steam  pressure  was  better  maintained  in  the  extended 
smoke-box  engine. 

The  shape  of  the  chimney  has  also  an  important 
bearing  on  the  character  of  the  blast.  This  has 


i]  STEAM  GENERATION  27 

formed  the  subject  of  experimental  determination 
at  the  Purdue  University,  U.S.A.,  by  Professor  Goss 
following  on  those  conducted  in  1896  on  the  Continent 
by  Von  Borries.  The  outcome  of  these  experiments 
was  that  it  is  preferable  to  use  a  chimney  of  conical 
shape,  that  the  diameter  should  be  small  enough  to 


Fig.  11.     Diagram  shewing  the  piston  action  of  the  blast. 

cause  the  cone  of  exhaust  steam  to  strike  the  barrel 
of  the  chimney,  and  of  such  a  length  that  the  steam 
would  form  a  fluid  piston  capable  of  setting  up  by 
its  movement  a  sufficient  vacuum  in  the  smoke-box. 
To  effect  this,  the  taper  of  the  chimney  should  be 
such  that  the  puff  of  steam  continues  to  fit  until 


28  THE  MODERN  LOCOMOTIVE         [CH. 

it  finally  emerges  into  the  atmosphere  (Fig.  11). 
Practice  has  shewn,  particularly  in  the  case  of  certain 
British  locomotives  of  recent  construction,  that  a 
reduction  in  the  length  of  chimney  does  not  inter- 
fere sensibly  with  the  blast.  Professor  Goss  has 
shown  that  the  practice  of  prolonging  the  chimney 
on  the  interior  side  of  the  smoke-box  below  the 
petticoat  pipe,  a  practice  in  vogue  on  several  Ameri- 
can and  European  lines,  has  caused  a  diminution  in 
the  effect  of  the  blast,  and  it  is  therefore  preferable 
to  abandon  it. 

Recent  practice,  as  we  shall  see  in  another  chapter, 
has  adopted  a  blast  pipe  with  a  variable  nozzle 
(A,  Fig.  10),  the  orifice  of  which  can  be  adjusted  to 
the  demand  for  steam.  American  engineers  have 
long  since .  recognized  the  theoretical  superiority  of 
this  arrangement,  and  its  adoption  has  also  found 
favour  in  Europe.  Some  difficulty  appears  to  have 
been  found  in  maintaining  the  parts  in  working  order, 
but  this  is  not  a  serious  matter. 

The  smoke-box  contains  also  the  spark  arrester 
(S,  Fig.  9),  by  which  the  throwing  of  live  coal  is 
diminished,  and  the  dead  cinders  are  kept  in  such 
a  position  in  the  smoke-box  as  to  remain  clear  of 
the  bottom  rows  of  tubes.  The  pattern  used  by 
Mr  J.  F.  Mclntosh  on  the  Caledonian  Railway, 
consists  of  a  V-plate  interposed  between  the  smoke- 
box  tube  plate  and  the  blast-pipe,  and  extends  from 


i]  STEAM  GENERATION  29 

the  bottom  of  the  smoke-box  to  a  level  just  above 
the  top  row  of  tubes.  This  deflector  plate  is  pivoted 
on  brackets  at  the  back  of  the  blast-pipe,  thus 
allowing  the  plate  to  be  swung  round  and  the  tubes 
to  be  cleaned. 

In  working,  the  ashes  drawn  through  the  tubes 
are  deflected  away  from  the  strong  current  caused  by 
the  blast,  and  instead  of  settling  down  gradually  and 
blocking  the  lower  tubes,  the  cinders  roll  back  to 
within  the  V  of  the  deflector,  being  thus  kept  away 
from  the  tubes. 

The  cylindrical  shell  of  the  boiler  is  lagged,  i.e. 
covered  with  some  non-conducting  material  to  pre- 
vent loss  of  heat  by  radiation  and  to  impart  a  finished 
appearance.  Wood  and  felt  with  an  outer  covering 
of  Russian  iron  were  extensively  used,  but  when 
running  the  wood  would  frequently  catch  fire,  neces- 
sitating the  stoppage  of  the  engine  to  extinguish  it. 

The  writer  once  had  a  very  unpleasant  footplate 
experience  on  the  Great  Central  Railway.  On 
entering  the  Woodhead  tunnel,  which  is  one  of  the 
longest  in  the  country  and  only  of  sufficient  width 
to  accommodate  one  train,  the  lagging  fired  with 
the  formation  of  clouds  of  dense,  suffocating  smoke, 
which  enveloped  the  footplate  and  engine.  In  the 
confined  space  of  the  tunnel  the  conditions  are  better 
imagined  than  described,  especially  as  it  was  im- 
possible to  stop  until  the  open  air  was  again  reached. 


30  THE  MODERN  LOCOMOTIVE         [CH. 

Modern  engines  are  fitted  with  asbestos  coverings 
and  an  outer  envelope  of  sheet  steel. 

Boiler  fittings  and  accessories,  with  the  exception 
of  the  injector,  which  will  be  dealt  with  elsewhere, 
do  not  concern  us,  since  they  play  no  part  in  steam 
raising. 

Having  now  obtained  a  general  idea  of  the  con- 
struction of  the  boiler,  we  shall  in  a  succeeding 
chapter  examine  some  of  the  modern  developments 
introduced  to  improve  its  working.  To  appreciate 
the  significance  of  these,  some  attention  must  now 
be  given  to  combustion  and  vaporization. 


CHAPTER  II 

COMBUSTION   AND  VAPORIZATION 

FUELS  depend  for  their  heating  value  upon  the 
presence  of  such  calorific  constitutents  as  carbon, 
hydrogen,  and  compounds  of  those  bodies  called 
hydrocarbons.  The  chemical  union  of  these  with 
oxygen  forms  the  familiar  process  of  combustion. 
It  is  accompanied  by  the  evolution  of  light  and 
heat,  which  latter  is  transferred  to  the  water  in 
the  boiler  in  the  way  we  have  seen. 

The   principal   chemical    combinations   resulting 


n]      COMBUSTION  AND  VAPORIZATION      31 

from  the  combustion  of  hydrogen,  carbon,  oxygen, 
are  carbon  monoxide  (CO)  and  carbon  dioxide  or 
carbonic  acid  gas  (C02).  The  former  results  from  the 
partial  combustion  of  carbon  with  a  limited  supply 
of  oxygen,  the  latter  by  perfect  combustion  secured 
by  a  copious  supply  of  oxygen.  Two  important  hydro- 
carbons are  also  formed  in  gaseous  form,  namely, 
methane  or  marsh  gas  (CH4)  and  ethylene  or  olefiant 
gas  (CH2).  In  the  locomotive  the  necessary  oxygen  is 
obtained  from  atmospheric  air  which  contains  oxygen 
and  nitrogen  in  the  proportion,  roughly,  of  one  part 
of  the  former  to  four  parts  of  the  latter  by  volume, 
or  56  parts  of  nitrogen  to  16  parts  of  oxygen  by 
weight.  Thus,  to  obtain  one  cubic  foot  of  oxygen, 
it  is  necessary  to  supply  five  cubic  feet  of  air ;  for 

one  pound  of  oxygen,  we  must  have  — — — =  4£  Ibs. 

of  air  roughly.  Nitrogen,  for  the  purposes  of  com- 
bustion, is  inert.  It  becomes  heated,  however,  in  the 
process  and  so  absorbs  a  considerable  proportion  of 
the  heat  developed  in  the  fire-box,  thus  limiting  the 
maximum  temperature  obtained.  The  heat  evolved, 
or  the  heating  value,  depends  upon  the  relative  pro- 
portions of  the  constituents  and  has  been  determined 
for  the  separate  elements  as  follows:  one  Ib.  of 
hydrogen  when  burnt  with  oxygen  to  form  water 
evolves  62,032  British  Thermal  Units  (usually  ex- 
pressed B.T.U.),  which  suffices  to  evaporate  64*2  Ibs. 


32  THE  MODERN  LOCOMOTIVE         [CH. 

of  water  from  and  at  212°  F.*  It  requires  for  its 
combustion  8  Ibs.  of  oxygen.  One  pound  of  carbon 
when  completely  burnt  evolves  14,500  B.T.U.  which  is 
sufficient  to  evaporate  15  Ibs.  of  water  from  and  at 
212°F.  For  combustion  2f  Ibs.  of  oxygen  are  required. 
When  partially  burned,  and  carbon  monoxide  is 
formed,  only  4400  heat  units  are  evolved  capable  of 
evaporating  4*55  Ibs.  of  water  from  and  at  212°  F. 
For  its  combustion  If  Ibs.  of  oxygen  are  required. 

The  fuel  in  most  common  use  for  locomotives  is 
coal.  Mineral  oil  is  used  to  some  extent  on  the  Gt 
Eastern  Railway,  and  more  extensively  in  Russia  and 
America.  Coke  was  formerly  employed  exclusively 
owing  to  its  smokeless  combustion,  and  it  was  not 
until  the  invention  of  the  brick  fire-arch  that  coal- 
burning  was  rendered  possible. 

The  principal  varieties  of  coal  found  in  this  country 
are  anthracite,  semi-anthracite,  and  semi-bituminous. 
The  two  last  mentioned  are  used  chiefly  for  steam 
raising,  the  best  being  the  Welsh  steam  coals,  which 
burn  readily  without  the  formation  of  black  smoke. 

*  It  is  usual  in  expressing  evaporation  results  to  use  a  common 
standard,  namely,  that  of  the  number  of  pounds  of  water  at  a  tem- 
perature of  212°  F.,  which  would  be  converted  into  steam  of  the  same 
temperature  by  the  application  of  the  same  number  of  heat  units. 
Each  pound  of  water  so  evaporated  would  take  up  966  B.T.U. ,  hence 

heating  value  in  B.T.U.  per  Ib. 
we  get  equivalent  evaporation  =  —  — ^^ —  —  . 


n]      COMBUSTION   AND  VAPORIZATION      33 

The  following  table  gives  the  heating  values  of  the 
principal  varieties  of  fuel. 


Fuel 

Heat  of  Combustion  in 
Thermal  Units  per  Ib. 

Best  Welsh  Coal  

15,000—16,000 

Newcastle    ... 

14,820 

Derbyshire  and  Yorkshire 

13,860 

Lancashire             

13,900 

Scotch         

14,100 

Coke            

12,500 

Mineral  Oil 

19,000—19,500 

These  are  theoretical  values  calculated  from  the 
chemical  composition,  and  upon  the  assumption  that 
one  pound  of  pure  carbon  is  capable  of  evaporating 
1 5  Ibs.  of  water  from  2 1 2°  F.  But,  as  we  shall  presently 
see,  the  practical  results  differ  considerably  from  the 
theoretical.  Coal,  however,  is  not  usually  bought  by 
railway  companies  on  a  basis  of  chemical  constituents, 
although  tests  for  heating  values  are  regularly  made 
for  checking  purposes.  These  values  may  be  calcu- 
lated from  the  composition  as  found  by  chemical 
analysis  or  determined  by  means  of  a  calorimeter, 
the  latter  method  giving  perhaps  the  more  certain 
results.  This  is  not  the  place  to  examine  the  phe- 
nomena of  thermal  changes  which  accompany  chemical 
reactions,  but  in  estimating  calorific  values  they  are 
of  importance.  For,  just  as  chemical  union  may  be 

A.   L.  3 


34  THE  MODERN  LOCOMOTIVE          [OH. 

accompanied  by  the  evolution  of  heat,  so  a  corre- 
sponding dissociation  requires  its  expenditure  and 
must  be  taken  into  account.  Again,  the  constituents 
may  combine  to  form  compounds  other  than  the 
products  of  combustion  as,  for  example,  where  a  fuel 
contains  both  oxygen  and  hydrogen,  some  of  the 
hydrogen  will  combine  with  some  of  the  oxygen  to 
form  water,  consequently  no  heat  will  be  available 
from  this  hydrogen.  Thus  the  weight  of  hydrogen 
available  in  1  Ib.  of  coal  for  calculating  the  heating 

value  will  be  given  by  H  -  — ,  since  8  parts  by  weight 

o 

of  oxygen  unite  with  one  by  weight  of  hydrogen  to 
form  water. 

Determination  of  the  calorific  value  by  the  calori- 
meter may  be  made  by  a  Mahler  apparatus,  which 
consists  of  a  steel  shell  containing  a  pound  of  com- 
bustible. This  is  ignited  by  an  electric  spark  and 
is  burnt  instantaneously  by  the  aid  of  pure  oxygen 
introduced  under  a  pressure  of  400  Ibs.  per  square 
inch.  Before  ignition,  the  shell  is  immersed  in  a 
water  calorimeter. 

We  have  said  that  the  practical  heating  power 
of  coal  differs  from  its  theoretical  calculated  value. 
This  is  accounted  for,  first,  by  the  waste  of  fuel  and, 
secondly,  the  inability  of  the  boiler  to  utilize  all  the 
heat  generated  in  the  fire-box.  Taking  the  last  men- 
tioned cause  first,  the  evaporation  which  takes  place 


ii]      COMBUSTION  AND  VAPORIZATION      35 

in  a  boiler  is  only  about  7  to  8  Ibs.  of  water  per  Ib. 
of  coal,  representing  about  9J  Ibs.  from  and  at  212°  F. 
and  a  boiler  efficiency  of  probably  about  66  per  cent. 
This  low  evaporative  duty  is  chiefly  due  to  the  high 
temperature  retained  by  the  gases  after  they  leave 
the  tubes,  about  25  per  cent,  of  the  available  heat 
being  wasted  in  this  manner.  Part  of  this  loss  is 
unavoidable,  as  it  is  impossible  for  the  gases  to  cool 
down  to  the  temperature  of  the  steam,  some  head  of 
temperature  being  necessary  to  enable  the  heat  to 
penetrate  the  metal. 

Unskilful  stoking  is  also  a  source  of  waste,  so 
much  so,  that  it  is  the  practice  on  most  lines  to  give 
bonuses  for  coal  saving.  Excessive  coaling  means 
that  sufficient  air  cannot  reach  a  portion  of  the  fire, 
hence  some  of  the  coal  being  only  warmed  will  be 
distilled  and  part  with  its  valuable  and  volatile  hydro- 
carbon constituents  in  the  shape  of  unburnt  gas;  a 
proportion  of  the  remainder  will  be  incompletely 
burnt  to  carbon  monoxide,  which  means  that  only 
4400  heat  units  per  pound  of  carbon  are  generated 
instead  of  14,500. 

Much,  however,  can  be  done  by  proper  manipu- 
lation of  the  damper  and  firehole  door,  and  a  good 
fireman  will  be  influenced  by  his  position  on  the 
road  when  firing  up.  Weather  conditions  also  exert 
a  great  influence  on  the  fireman's  work,  an  engine 
being  generally  found  to  steam  best  against  a  head 

3—2 


36  THE  MODERN  LOCOMOTIVE          [OH. 

wind,  and  worst  with  a  side  wind ;  in  the  former  case 
the  air  is  forced  well  into  the  ashpan  and  through  the 
fire  bars,  whilst  in  the  latter  case  the  wind  rushing 
across  underneath  the  engine  has  a  tendency  to  suck 
out  the  air  in  the  ashpan,  acting  much  as  a  steam 
ejector  would  do. 

A  considerable  quantity  of  small  coal  is  drawn 
through  the  tubes  by  the  fierce  draught,  and  as 
most  of  it  is  in  an  incandescent  state,  an  appreciable 
loss  occurs.  The  loss  is  greatest,  of  course,  when 
using  the  small  coal  commonly  known  as  'slack,'  and 
with  Welsh  coal,  which  splits  up  into  small  pieces 
when  heated,  instead  of  caking  together  in  a  pasty 
mass  like  the  bituminous  varieties.  In  recent  years 
spark-throwing  has  been  much  diminished,  as  we  have 
seen,  through  the  introduction  of  spark  arresters.  A 
fierce  blast  is  also  unfavourable  to  coal  economy,  not 
only  because  it  tends  to  increase  spark-throwing,  but 
the  gases  are  drawn  through  the  tubes  at  such  a  high 
velocity  that  they  have  not  time  to  give  up  their  heat 
and  the  smoke-box  temperature  rises  to  an  excessive 
degree.  Other  causes  of  inefficiency  are  contraction 
of  the  flue- way  area  of  tubes  by  ferules  or  otherwise, 
and  the  formation  of  a  non-conducting  deposit  or 
scale  due  to  the  presence  in  the  water  of  carbonates 
and  sulphates  of  lime. 

The  available  information  concerning  the  amount 
of  heat  lost  in  the  working  of  the  locomotive  has 


n]      COMBUSTION  AND  VAPORIZATION      37 

recently  been  added  to  by  the  results  obtained  by  the 
Breslau  Royal  Railway  Department  from  a  definite 
series  of  vaporization  tests.  It  is  worth  while  stating 
the  results  reached.  (1)  The  heat  escaping  in  the 
smoke  gases  was  20  to  23  per  cent,  of  the  total  heat 
value  of  the  coal.  (2)  The  loss  by  the  combustible 
components  found  in  the  residue  varied,  according  to 
the  design  of  the  locomotive,  from  5  to  11  per  cent, 
of  the  total  heat  value  of  the  coal.  (3)  The  loss  by 
radiation,  smoke  and  spark  production  was  about 
5  per  cent.  (4)  The  efficiency  of  the  boiler  there- 
fore varied  from  60  to  70  per  cent.  (5)  The  average 
temperature  in  the  smoke- box  was  from  330°  to 
380°  C.  (716°  F.).  (6)  The  mean  temperature  of 
combustion  in  the  fire-box  was  found  to  be  1485°  C. 
(2705°  F.).  (7)  The  mean  specific  heat  of  the  smoke 
gases  of  average  composition  reduced  to  0°  C.  is  0*324 
for  a  smoke-box  temperature  ranging  from  350 — 
400°  C.  (662—752°  F.)  and  0'35  for  the  fire-box  and 
tube  temperatures.  (8)  The  gases  with  the  loco- 
motive in  full  running  and  350°  C.  (662°  F.)  in 
the  smoke-box  were  found  to  contain  an  average 
of  11  per  cent,  carbon  dioxide  and  0*6  per  cent, 
carbon  monoxide.  From  one  kg.  (2*2  Ibs.)  of  coal 
11  to  12  cub.  ms.  (388—423  cub.  ft.)  of  smoke  gas 
were  obtained  at  0"C. 

The  inability  of  the  boiler  to  utilize  all  the  heat 
generated  from  the  fuel  follows  from  the  nature  of 


38  THE  MODERN  LOCOMOTIVE         [OH. 

heat  transmission  to  the  water.  The  heat  evolved  in 
the  fire-box  is  propagated  in  two  ways:  by  direct 
conduction  and  by  radiation.  By  the  first  the  heat  is 
usually  considered  as  propagated  by  the  hotter  mole- 
cules heating  the  neighbouring  colder  molecules  of 
the  plates;  by  the  second  the  transference  takes 
place  without  the  intervention  of  matter  by  etheric 
waves  set  up  by  the  vibration  of  the  heated  molecules. 
It  is  in  this  manner  that  radiant  heat  (and  light) 
reaches  the  earth  from  the  sun. 

If  there  were  no  radiant  heat  the  temperature 
of  the  fire-box  gases  would  be  extremely  high,  and 
all  the  heat  evolved  would  be  available  for  heating 
them.  As  great  a  difference  of  temperature  as  1800°  C. 
(3272°  F.)  can  exist  between  the  calculated  and  ascer- 
tained temperatures  in  the  fire-box,  which  affords  a 
measure  of  the  losses  due  to  the  radiant  heat. 

The  rate  of  conductivity  of  metal  is  such  that  a 
few  degrees'  difference  in  temperature  on  each  side 
of  a  plate  is  sufficient  to  account  for  the  transmission 
of  a  large  quantity  of  heat.  Thus  it  has  been  found 
that  a  vaporization  of  200  kgs.  per  sq.  m.  (41  Ibs.  per 
sq.  ft.)  per  hour  in  a  copper  fire-box  corresponds  to 
a  difference  of  only  47°  C.  (40'4°  F.)  between  the  two 
faces  of  a  plate  13  mm.  (|f  in.)  thick.  M.  Nadal, 
locomotive  engineer  of  the  French  State  Railways, 
has  stated  that  in  steel  tubes  of  2*5  mm.  (^  in.) 
thickness  a  vaporization  of  60  kgs.  per  sq.  m.  per 


n]      COMBUSTION  AND  VAPORIZATION      39 

hour  (12*3  Ibs.  per  sq.  ft.)  means  a  corresponding 
difference  in  temperature  of  only  17°  C.  (35°  F.). 

From  the  hot  gases  to  the  fire-box  walls,  and 
from  the  latter  to  the  water,  heat  is  transmitted  by 
external  conduction.  The  coefficient  of  conductivity 
between  the  plates  and  the  water  is  high,  and  the 
temperature  difference  small ;  in  fact  the  temperature 
at  the  surface  of  the  plate  is  at  the  most  15 — 20°  C. 
(59—68°  F.)  higher  than  that  of  the  water.  On  the 
contrary,  the  coefficient  of  conductivity  between 
the  gases  and  the  plates  is  small,  which  necessitates 
increasing  the  contact  surface  to  the  greatest  possible 
extent.  Hence  we  get  the  Serve  type  of  tube.  It 
is  owing  to  the  high  conductivity  of  the  metal  and 
the  high  coefficient  of  exterior  conductivity  between 
the  metal  and  water  that  the  fire-box  surfaces  can 
readily  absorb  the  radiant  heat. 

It  follows,  therefore,  that  the  direct  heating  surface 
obtained  from  the  fire-box  and  portions  of  the  ad- 
jacent tube  produces  a  more  effective  evaporation 
than  the  rest  of  the  flue  area  and  accounts  for  from 
one-third  to  one-half  of  the  total  quantity  of  steam 
generated. 

The  power  developed  by  the  boiler  is  therefore 
in  reality  limited  by  the  grate  area  and  the  maximum 
amount  of  coal  consumption.  With  a  given  grate 
area  and  a  given  strength  of  draught,  a  definite 
quantity  of  coal  can  be  burnt  per  hour.  A  large 


40  THE  MODERN  LOCOMOTIVE          [CH. 

grate  area  means  that  a  large  amount  of  heating 
surface  must  be  provided  to  ensure  the  efficient  utili- 
zation of  the  heat.  Thus  the  evaporative  power  of 
the  boiler  depends  upon  the  ratio  of  heating  surface 
(HS)  to  grate  area  (GA)  and  rate  of  coal  consumption. 
It  is  generally  stated  in  pounds  of  water  evaporated 
from  feed  temperature  per  square  foot  of  heating 
surface  per  hour. 

TTO 

The  ratio          varies  between   60  and  100,  the 


average  being  about  80.  With  a  ratio  of  less  than  60 
the  flue  area  will  be  reduced  to  require  a  very  sharp 
blast  which  we  have  seen  to  be  a  disadvantage,  while 
increasing  it  to  over  100  means  crowding  of  tubes 
or  undue  elongation  of  them.  The  former  obstructs 
the  water  circulation,  and  the  latter  means  increased 
frictional  resistance  to  the  flow  of  the  hot  gases  and 
much  reduced  temperature  in  the  last  foot  or  two  of 
length,  which  becomes  then  of  little  value  as  heating 
surface. 

The  rate  of  coal  consumption  reaches  150  Ibs.  or 
more  per  square  foot  of  grate  area  per  hour  for  short 
periods  and  fast  running  ;  it  may  average  on  a  run 
with  a  train  load  of,  say,  300  tons,  90  Ibs.  per  sq.  ft. 
per  hour. 

The  amount  of  water  evaporated  averages  30  Ibs. 
per  indicated  horse-power  per  hour,  measured  from 
the  tender,  of  which  at  least  21  Ibs.  are  required 


in]  BOILER  IMPROVEMENTS  41 

for  the  engine,  the  balance  being  consumed  for 
working  the  injector,  blower,  brakes,  and  by  blowing 
off  at  the  safety  valve.  Or,  it  may  be  stated  that  as 
much  as  13  Ibs.  of  water  can  be  evaporated  from  one 
sq.  ft.  of  heating  surface  per  hour.  An  average  of 
3  sq.  feet  of  total  heating  surface  per  indicated  horse- 
power may  be  taken  as  an  approximation. 

The  modern  big  boiler  has  another  advantage 
besides  that  due  to  the  large  grate  area  and  heating 
surface,  namely,  its  capacity  for  carrying  a  large 
volume  of  hot  water.  Thus,  should  the  steam  pres- 
sure shew  a  tendency  to  fall  when  nearing  the  top  of 
a  long  bank,  the  feed  can  be  shut  off,  thus  temporarily 
increasing  the  boiler  power  by  some  25  per  cent., 
owing  to  the  fact  that  the  latent  heat  of  evaporation 
only  has  to  be  supplied.  Some  three  or  four  miles 
can  be  run  with  the  feed  shut  off  without  letting  the 
water-level  drop  dangerously  low. 


CHAPTER   III 

INCREASING  THE  USEFUL  EFFECT  OF  THE  BOILER 

WITH  the  object  of  increasing  the  efficiency  of  the 
standard  boiler,  and  of  obtaining  increased  power 
from  it,  numerous  devices  and  experiments  have  been 
tried  in  recent  years,  some  of  which  have  yielded 


42  THE  MODERN  LOCOMOTIVE          [OH. 

sufficiently  satisfactory  results  to  justify  their  per- 
manent adoption.  Some  of  these  methods  such  as 
coning  the  boiler,  varying  the  fire-box  contour,  the 
employment  of  water  tubes,  pre-heating  the  feed 
water,  thermal  storage  and  superheating,  will  now  be 
examined. 

Cone  Boiler.  This  type  of  boiler,  an  example  of 
which  is  seen  in  the  Gt  Bear,  Fig.  3,  forms  a  prominent 
feature  in  recent  locomotives  of  the  Gt  Western 
Railway,  designed  by  Mr  Churchward.  Some  of  the 
boiler  rings,  instead  of  being  truly  cylindrical,  form 
the  frustum  of  a  cone,  with  the  result  that  the 
largest  cross-sectional  area  of  the  boiler  barrel  is 
in  the  locality  where  we  have  seen  the  highest  in- 
tensity of  combustion  takes  place,  consequently  where 
the  heating  surface  is  the  most  valuable,  namely,  close 
to  the  smoke-box.  This  is  a  distinct  and  definite 
advantage,  since  it  provides  a  greater  area  of  water 
line,  an  increased  steam  capacity  and,  by  the  larger 
diameter  being  arranged  to  coincide  with  the  line  of 
the  fire-box  tube  plate,  much  more  water  space  at 
the  sides  of  the  tubes. 

It  has  also  materially  contributed  to  the  reduction 
of  priming  or  foaming  and  it  enabled  the  dome,  always 
a  source  of  weakness,  to  be  dispensed  with  and  at 
the  same  time  to  secure  dry  steam.  This  important 
result  has  also  been  obtained  by  the  employment  of  a 
fire-box  with  a  flat  top,  the  most  conspicuous  example 


in]  BOILER  IMPROVEMENTS  43 

of  which  is  the  invention  of  a  Belgian  engineer, 
M.  Belpaira 

Belpaire  F ire-Box  (Fig.  12).     In  this  the  outer 


Fig.  12.     Details  of  a  Belpaire  fire-box. 

wrapper  is  made  flat  on  top,  parallel  to  the  roof  of 
the  inner  fire-box,  to  which  it  is  tied  by  stays  similar 
to  those  used  for  connecting  the  sides  of  the  ordinary 


44  THE  MODERN  LOCOMOTIVE          [OH. 

fire-box.  The  first  two  rows  only  are  provided  with 
sling  stays  for  securing  vertical  flexibility  to  the 
front  of  the  box,  and  cross-stays  connect  the  sides 
of  the  walls  of  the  outer  wrapper  above  the  inner 
box.  With  the  flat  top  the  area  of  the  water  line  at 
the  hottest  part  of  the  boiler  is  increased  and  more 
steam  space  provided,  since  the  girder  stays,  which 
take  up  a  large  portion  of  the  heating  surface  of 
the  top  of  the  ordinary  box,  and  are  thought  by 
some  to  hinder  the  free  rising  of  the  steam  bubbles, 
are  dispensed  with.  Mr  Churchward  states  that  less 
trouble  has  been  experienced  on  the  Gt  Western 
Railway  with  the  Belpaire  box  than  with  the  round 
top.  It  is  therefore  not  surprising  to  find  that  it  is 
increasing  in  favour,  and  to-day  it  is  employed  by  the 
Gt  Central,  Midland,  North  British,  Lancashire  and 
Yorkshire,  Gt  Eastern,  the  last  to  adopt  it  being 
the  London  and  North  Western. 

W ootten  Fire- Box  (Fig.  13).  The  deep,  round- 
topped  fire-box  spreading  wide  outside  the  frames — 
a  feature  of  Mr  Ivatt's  famous  Atlantic  engines  on 
the  Gt  Northern  Railway — is  known  as  the  Wootten 
fire-box.  Its  employment  is  possible  only  when  the 
rear  pair  of  coupled  wheels  are  set  well  forward  and 
a  pair  of  trailing  wheels  used  to  carry  it.  Hence 
it  is  in  favour  on  the  4-4-2  and  4-6-2  types  (see 
p.  123).  The  lateral  walls  are  strongly  inclined  towards 
the  exterior,  for  example  with  a  barrel  of  5  ft.  8  in. 


Ill] 


BOILER   IMPROVEMENTS 


45 


diameter,  the  width  of  the  base  of  the  fire-box 
reaches  as  much  as  7  ft  2  in.  The  Wootten  box 
originated  in  America,  where  it  is  still  largely  used. 
Designed,  however,  to  burn  inferior  fuel,  the  recog- 
nized advantage  of  employing  high  class  coal  has  led 
to  a  marked  development  with  the  object  of  reducing 


Fig.  13.     Details  of  a  Wootten  fire-box. 

its  solidity  and  cost  of  upkeep.  Moreover,  the  in- 
clination of  the  lateral  walls,  if  carried  too  far,  retards 
the  ascending  currents  of  steam  since  the  course  of 
circulation  in  a  boiler  is  upwards  in  the  fire-box,  and 
downwards  in  the  smoke-box  end. 


46 


THE  MODERN  LOCOMOTIVE 


[CH. 


American  Fire-Box.  Fig.  14  represents  one  of 
the  most  recent  types  of  large  American  fire-boxes. 
It  is  simple  in  form,  with  straight  lateral  walls  and 
sloped  back  plate.  The  lower  part  of  the  front  tube 
plate  is  also  sloped,  which  allows  the  box  proper 


Fig.  14.     American  type  tire-box. 

to  be  prolonged  into  the  barrel  for  about  3  ft.,  so 
constituting  a  combustion  chamber  increasing  to  a 
certain  extent  the  evaporative  efficiency.  American 
fire-boxes  constructed  without  spread  of  the  lateral 
walls  are  called  'wagon  top/ 


Ill] 


BOILER  IMPROVEMENTS 


47 


Jacobs-Schupert  Fire-Box  (Fig.  15).  In  this,  the 
latest  American  development,  the  usual  arrangement 
of  flat  plates  supported  by  stay  bolts  has  been  aban- 
doned, except  in  the  front  and  back  sheets.  Side 
and  wrapper  plates  have  been  replaced  by  channel- 
shaped  sections  riveted  together.  These  are  stayed 
by  stay  sheets  interposed  between  the  sections.  All 
seams  are  submerged,  and  no  joints  are  exposed  to 
the  direct  currents  of  heat  and  gases.  Owing  to  the 


Fig.  15.     Jacobs-Schupert  fire-box. 

irregular  outline  thus  formed  for  the  crown  and 
sides,  the  available  heating  surface  of  the  hottest 
section  of  the  boiler  is  enlarged  Avithout  increasing 
the  size  of  the  grate  area,  and  the  arched  concave 
construction  of  the  sections  ensures  that  there  will 
be  no  undue  local  stresses,  the  shape  of  each  section 
being  such  that  it  will  expand  or  contract  with 
variations  in  temperature  and  produce  only  small 
stresses  on  adjacent  sections. 


48  THE  MODERN  LOCOMOTIVE          [CH. 

Stayless  Boiler.  A  number  of  attempts  have 
been  made  in  Europe  to  dispense  altogether  with 
the  fire-box.  Herr  Lenz  in  Germany  some  years 
ago  introduced  a  corrugated  form  of  fire-box  which 
was  claimed  to  be  sufficient  in  itself  to  support  the 
tube  plates,  and  no  further  stays  were  used.  After 
some  disastrous  explosions,  however,  they  were  with- 
drawn from  service.  Vanderbilt,  on  the  Prussian 
Railway,  has  also  employed  a  stayless  boiler  with  a 
corrugated  fire-box. 

Water-tube  Boilers  and  Fire-Boxes.  In  spite  of 
the  extensive  adoption,  within  recent  years,  of  the 
water-tube  type  of  boiler  for  both  land  and  marine 
service,  little  has  been  heard  of  the  possibilities  of 
this  type  of  steam  generator  in  connection  with  the 
railway  locomotive.  This  is  more  particularly  striking 
in  view  of  the  quick  steaming  requirements  of  the 
modern  locomotive  and  the  special  qualities  which 
appear  to  be  possessed  by  the  water-tube  boiler  for 
meeting  them. 

The  apparent  diffidence  with  which  the  water- 
tube  boiler  problem  has  been  treated  by  locomotive 
engineers  has,  however,  met  with  some  notable 
exceptions  in  the  case  of  Herr  Brotan,  the  cele- 
brated Austrian  engineer,  and  Mr  D.  Drummond, 
the  Locomotive  Superintendent  of  the  London  and 
South  Western  Railway,  who  for  some  years  past  have 
consistently  made  use  of  water-tubes  with  a  great 


Ill] 


BOILER  IMPROVEMENTS 


49 


amount  of  success.  The  feature  of  the  Brotan 
system  (Fig.  16)  is  the  replacement  of  the  ordinary 
inside  fire-box  by  a  system  of  water-tubes,  involving 
the  elimination  of  the  customary  water  space  and 
stays.  The  boiler  proper  is  divided  into  two  cylin- 
drical barrels  fixed  parallel  to  each  other,  the  lower 
and  main  portion  containing  a  number  of  fire-tubes. 
Connection  with  the  upper  barrel  is  made  by  means 


Fig.  16.     Brotan  water-tube  boiler. 

of  two  necks.  The  upper  ends  of  the  water-tubes, 
composing  the  fire-box,  are  fixed  in  the  rear  end 
of  the  upper  barrel,  which  is  therefore  of  thicker 
plate.  The  lower  ends  of  the  water-tubes  are  ex- 
panded into  a  rectangular  shaped  water-circulating 
chamber  of  steel,  connection  between  which  and  the 
back  end  of  the  main  barrel  is  made  by  a  large  pipe. 
Facilities  for  inspecting  and  cleaning  the  water-tubes 
are  provided  by  means  of  a  number  of  removable 

A.  L.  4 


50  THE  MODERN  LOCOMOTIVE          [CH. 

doors  on  the  underside  of  the  circulating  chamber, 
giving  access  to  corresponding  holes  into  which  the 
water-tubes  themselves  are  expanded.  The  fire-box 
tubes  are  encased  in  fire-clay  and  a  covering  of  sheet 
steel  plates.  One  example  of  this  type  of  boiler, 
constructed  by  Messrs  Beyer,  Peacock  &  Co.,  exists 
at  the  works  of  the  Mannesman  Tube  Company.  On 
the  Austrian  State  Railways,  they  have  been  employed 
with  notable  success  since  1901. 

Schneider  Boiler.  Somewhat  similar  in  type  was 
the  boiler  of  a  locomotive  shewn  at  the  recent  Nancy 
Exhibition  by  Messrs  Schneider  &  Cie  of  Creusot. 
The  boiler  consists  of  an  upper  drum,  containing 
water  and  steam,  extending  along  the  whole  length 
of  the  boiler  and  connected  by  small  diameter  tubes 
to  four  water  collectors.  The  rear  pair  of  the  latter 
are  placed  one  on  each  side  of  the  fire-box. 

Each  pair  of  front  and  back  collectors  are  inter- 
connected by  cast  steel  tubes,  and  communication 
between  these  and  the  upper  drum  is  made  by  a 
return  water-tube.  The  back  water-tubes  are  splayed 
to  enclose  the  grate  which,  together  with  the  tubes 
themselves,  forms  the  inside  fire-box.  The  tubes  are 
interlaced  at  the  top  to  screen  the  drum  from  the 
direct  action  of  the  flames,  and  the  tubes  in  each  of 
the  two  outside  rows  are  closely  juxtaposed  to  pre- 
vent the  escape  of  the  hot  gases.  The  front  and  back 
portions  of  the  fire-box  are  built  up  of  fire-brick, 


HI}  BOILER  IMPROVEMENTS  51 

which  material  is  also  used  as  a  covering  for  the 
lower  portions  of  the  water-tubes  and  collectors.  The 
front  tubes  are  disposed  so  as  to  form  a  horizontal 
flue  for  the  passage  of  the  products  of  combustion 
to  the  smoke-box,  the  tubes  being  arranged  in  rows 
in  a  longitudinal  direction.  As  in  the  case  of  the 
back  group,  escape  of  gases  from  the  sides  is  prevented 
by  close  contact  of  the  tubes  in  the  two  outer  rows. 

The  tubes  connecting  the  drum  and  collectors  are 
inlet  on  the  underside  of  the  drum,  and  a  very  low 
level  of  water  suffices  to  cover  them  entirely.  This 
is  held  to  constitute  an  important  advantage  peculiar 
to  this  type  of  boiler,  as  the  volume  of  free  water 
comprised  within  the  maximum  and  minimum  levels 
is  sensibly  greater  than  that  which  is  available  within 
the  same  limits  in  cylindrical  boilers  of  the  ordinary 
smoke-tube  type.  This  means  a  greater  reserve  of 
energy,  which  can  be  drawn  upon  when  long  gradients 
have  to  be  negotiated. 

The  surface  of  ebullition  remains  nearly  constant 
whatever  may  be  the  height  of  the  water-level  in  the 
drum,  because  it  is  always  in  the  neighbourhood  of 
the  horizontal  diameter  of  the  drum.  Such  is  not  the 
case  in  boilers  of  the  ordinary  type,  in  which  the  sur- 
face of  ebullition  diminishes  progressively  with  the 
height  of  water-level,  so  that  priming,  which  results 
from  such  a  diminution  of  evaporating  surface,  cannot 
take  place  in  the  new  type  of  boiler.  The  outside 

4—2 


52  THE  MODERN  LOCOMOTIVE         [CH. 

covering  is  made  up  of  removable  segments,  pro- 
longed at  the  front  end  to  form  a  smoke-box. 

Riegel  on  the  Southern  Pacific  Railroad  of  America 
uses  a  water-tube  boiler  on  express  passenger  engines. 
The  water-tubes  are  located  in  the  fire-box,  and  the 
foundation  ring,  which  is  of  cast  steel,  has  water- 
pockets  cast  in  it  at  the  sides,  beyond  the  grate  and 
throughout  its  length,  thus  forming  lower  termina- 
tions for  two  nests  of  water-tubes.  These  extend 
from  the  pockets  diagonally  upwards  to  the  crown 
plate,  which  is  slightly  depressed  to  keep  the  upper 
tube  terminations  flooded.  Above  the  crown  plate 
is  provided  a  staying  cylinder,  which,  with  the  crown 
plate,  makes  a  double  thickness  at  the  crown  for 
tube  ends ;  this  cylinder  has  sufficient  flexibility  to 
allow  for  expansion  and  contraction.  The  tubes  can 
be  withdrawn  through  the  water-pockets  which  are 
fitted  with  removable  plates.  This  fire-box  has  no 
less  than  768  sq.  feet  of  heating  surface. 

Marine  Type.  The  marine  type  of  water-tube 
fire-box  (Fig.  17)  employed  with  satisfactory  results 
on  the  Northern  of  France  Railway  deserves  mention. 
An  engine  so  fitted  was  shewn  at  the  1910  Brussels 
Exhibition  after  having  covered  33,000  kms.  on  the 
road.  In  vertical  cross-section  the  fire-box  resembles 
the  Wootten  overhanging  type,  affording  accommoda- 
tion for  a  group  of  splayed  water-tubes  which  form 
the  side  walls  of  the  box.  The  tubes  are  expanded, 


Ill] 


BOILER  IMPROVEMENTS 


53 


in  the  manner  peculiar  to  marine  practice,  into  a 
header  or  cylinder  at  the  top  of  the  box  and,  at 
the  bottom,  into  two  water  legs  or  drums  extending 
laterally  along  the  sides  of  the  fire-box.  The  high 
pressure  of  255  Ibs.  per  sq.  in.  is  employed. 


Fig.  17. 


Marine  type  of  water-tube  fire-box  ;  Northern  of  France 
Eailway. 


A  fire-box  of  this  type  has  been  fitted  to  one  of 
the  new  Baltic  4-6-4  type  engines  of  the  Northern 
of  France  Railway,  illustrated  in  Fig.  1 9. 

It  is,  however,  Mr  Drummond  who  has  made  the 
most  consistent  use  of  the  water-tube  principle,  and 
practically  all  the  engines  on  the  London  and  South 
Western  Railway  are  so  fitted. 


Fig.  18.     4-6-0  type  4-cylinder  simple  expansion  engine,  with 
water-tube  fire-box ;  London  and  South  Western  Railway. 


Fig.  19.     4-6-4  ('Baltic)  type  4-cylinder  compound  express  locomotive  ; 
Northern  of  France  Eailway. 


Fig.  20.     4-4-2  (Atlantic)  type  express  locomotive  ; 
North  British  Railway. 


CH.  in]        BOILER   IMPROVEMENTS 


55 


They  have  proved  themselves  to  be  more  eco- 
nomical in  coal  consumption  than  similar  engines 
fitted  solely  with  flue  tubes.  This  is  doubtless  due 
to  the  direct  cycles  of  water  circulation  through  the 
tubes  and  about  the  fire-box  in  general,  which  cause 


Fig.  21.     Water-tube  fire-box  ;  London  and  South  Western  Railway. 

rapid  heat  absorption  and  prevent  scale  formation. 
The  writer  can  testify  to  the  remarkable  absence  of 
scale  as  a  result  of  an  inspection  of  these  engines 
immediately  after  being  taken  off'  the  road  for  repairs. 
Transverse  water-tubes  are  employed  as  shewn  in 
Fig.  21.  They  are  of  mild  steel,  slightly  inclined 


56  THE  MODERN  LOCOMOTIVE          [OH. 

and  rolled  into  the  lateral  walls  of  the  inner  fire- 
box. Transverse  stays  are  passed  through  some  of 
the  tubes  to  stiffen  the  box  suitably.  Access  to  the 
tubes  is  obtained  by  a  hinged  door  at  the  side, 
accurately  faced  to  form  a  steam-tight  joint  with 
the  faced  rectangular  castings  on  the  outer  fire-box. 
They  form  a  fine  example  of  hand  filing  and  of  a 
metal-to-metal  joint.  Incidentally  the  illustration 
shews  a  method  of  slinging  the  inner  fire-box  without 
the  use  of  girder  stays.  The  sling  bolts  are  in  couples 
with  nuts  bedding  on  crosspieces,  leaving  the  nuts 
free  to  lift  up  when  the  fire-box  rises  by  expansion. 
The  rising  pressure,  however,  brings  the  nuts  back 
again  on  their  seating. 

The  success  of  this  well-tested  water-tube  ar- 
rangement has  led  Mr  Drummond  to  pursue  his 
investigations  further,  for  which  purpose  he  built, 
some  time  ago,  a  locomotive  entirely  on  the  water- 
tube  principle.  The  results  obtained  with  this  are 
not  yet  known. 

Water  Softening.  The  incrustation  deposited  on 
the  walls  of  the  tubes  by  the  use  of  hard  waters,  i.e. 
water  containing  carbonate  and  sulphate  of  lime,  and 
magnesia  in  solution,  is  an  extremely  bad  conductor 
of  heat  and  its  presence  in  any  quantity  needs  not 
only  more  heat  to  evaporate  it,  but  leads  to  over- 
heating or  burning  of  the  plates.  The  trouble  from 
this  cause  increases  as  the  pressure  and  temperature 


in]  BOILER  IMPROVEMENTS  57 

of  the  steam  rise.  In  some  cases  it  has  been  found 
that  water  which  gave  little  or  no  trouble  at  160  Ibs. 
pressure  was  practically  unusable  at  200  Ibs.  More 
attention  is  now  paid  to  the  treatment  of  water  before 
it  is  used,  and  large  water-softening  plants  have  been 
installed  in  districts  where  the  water  is  notoriously 
hard.  The  systems  used  differ  in  the  method  of 
adding  the  chemicals,  but  they  depend  essentially 
upon  the  principle  that  free  carbon  dioxide  (C02) 
assists  to  keep  the  lime  in  solution.  If  an  excess 
of  lime  be  now  added  to  the  water,  the  C02  is 
neutralized  and  the  whole  of  the  lime,  including 
that  originally  present  in  the  water,  is  thrown  down 
as  a  precipitate.  Soda  ash  is  also  employed. 

An  electrolytic  method  of  treatment  has  been 
experimented  with  in  America  which,  although  costly, 
reduced  the  incrusting  solids  from  40  grs.  to  6  grs. 
per  gallon. 

Briefly  stated,  the  process  consists  in  submerging 
aluminium  or  iron  plates  in  the  water  and  then 
passing  an  electric  current  through  the  plates  which 
are  connected  up  in  series.  The  plates  enter  into  solu- 
tion in  proportion  to  the  quantity  of  water  treated. 

Oil-burning  Apparatus.  Liquid  fuel  is  used  on 
the  Gt  Eastern  Railway  and  on  the  locomotives 
running  in  the  oil-field  countries  such  as  Southern 
Russia,  the  Far  East,  and  the  Southern  States  of 
America  and  Mexico.  On  the  Southern  Pacific 


58  THE   MODERN  LOCOMOTIVE          [CH. 

Railroad  alone,  nearly  1000  locomotives  are  of 
the  oil-burning  type.  In  these  engines  the  oil  is 
carried  in  tanks  built  to  fit  the  coal  space  in  the 
tender. 

The  burner  used  is  of  the  flat-jet  type  consisting 
of  a  flat  casting,  divided  longitudinally  by  a  partition 
over  which  the  oil  flows  as  it  is  admitted  to  the  upper 
cavity.  The  lower  cavity  receives  the  steam  for  the 
jet  which  strikes  the  oil  flowing  over  the  partition, 
spraying  it  into  the  furnace  which  has  refractory  fire- 
bricks built  in  on  the  lower  sides  to  prevent  the  oil 
blast  impinging  against  the  sheets.  The  aim  is  com- 
pletely to  atomize  or  break  up  the  oil  near  the  burner 
tip  in  order  that  it  may  be  immediately  vaporized. 
The  steam  for  atomizing  is  obtained  from  the  dome. 
Other  methods  obtain  atomization  with  compressed 
air  which,  however,  is  liable  to  produce  in  the  furnace 
a  more  intense  local  heat  than  is  desirable.  With 
the  steam  jet  the  oil  is  sprayed  and  broken  up  so 
as  to  allow  the  air  admitted  through  the  proper 
dampers  to  mix  and  the  oil  to  be  consumed  com- 
pletely without  damage  to  the  plates. 

The  best  evaporative  results  obtained  from  steam 
jet  burners  give  an  evaporation  of  13  Ibs.  of  water 
from  and  at  212°  F.  With  the  air-jet  burner  supplied 
with  heated  air  at  5  to  7  Ibs.  per  sq.  in.,  16  Ibs.  of 
water  can  be  evaporated.  Tests  made  on  oil-burning 
locomotives  shew  that  temperatures  ranging  from 


Ill 


BOILER   IMPROVEMENTS 


59 


2500  to  2750°  F.  are  obtained  with  the  steam  jet 
type. 

On  the  Gt  Eastern  Railway,  the  Holden  system  is 
employed  whereby  creosote  is  used  as  an  auxiliary  to 
the  ordinary  fire.  An  inner  steam  jet  and  an  outer 


,8 


Fig.  22.     Latest  type  of  oil  burner  ;  Great  Eastern  Kailway. 

annular  jet  of  oil  spray  are  used,  which  play  over  a 
bed  of  incandescent  fuel. 

In  the  latest  form  of  the  Holden  steam  jet 
burner,  illustrated  by  Fig.  22,  the  spray  is  projected 
from  a  series  of  holes  D  arranged  at  a  slight  angle, 
so  that  the  streams  of  atomized  mixture  shall 


60  THE  MODERN  LOCOMOTIVE          [OH. 

converge  after  leaving  the  mixing  chamber.  Steam  is 
projected  from  a  series  of  holes  E,  and  supplied  by 
a  pipe  T  from  the  main  supply  entering  at  8 ;  the  oil, 
which  enters  at  slight  pressure  along  the  pipe  at 
the  side,  is  controlled  by  a  screwdown  valve  F,  in 
its  passage  to  the  base  of  the  outer  cone  N,  along 
which  it  is  drawn  by  an  annular  steam  jet  supplied 
at  about  60  Ibs.  pressure  to  the  inner  cone  K.  The 
steam  jet  also  induces  a  jet  of  air  from  A  by  way 
of  the  central  tube.  The  calorific  value  of  the  crude 
petroleum  used  varies  from  one  to  one  and  a  half 
times  that  of  coal.  Oil  fuel  has  also  been  employed 
on  the  engines  working  through  the  Arlberg  tunnel  on 
account  of  the  smokeless  combustion  of  the  fuel. 

Exhaust  Steam  Injectors.  The  apparatus  most 
generally  in  use  for  feeding  the  water  into  the  boiler 
is  the  Giffard  injector,  the  action  of  which  affords  one 
of  the  most  interesting  problems  in  thermo-dynamics. 
It  would  be  impossible  within  the  limits  of  this 
chapter,  to  examine  the  theory  of  its  working;  it 
must  suffice  to  state  that  it  depends  upon  a  rush  of 
steam  from  the  boiler  at  an  enormous  velocity  to 
induce  the  flow  of  a  corresponding  stream  of  cold 
water,  by  which  the  steam  is  condensed.  The  velocity 
attained  by  the  combined  stream  of  cold  water  and 
condensed  steam  is  sufficient  to  cause  it  to  enter  the 
boiler  against  the  same  internal  pressure  as  that  of 
the  steam  itself.  In  the  diagram  Fig.  23  steam  enters 


in]  BOILER  IMPROVEMENTS  61 

at  A  and  passes  through  the  nozzle  G.  Water  is 
drawn  in  at  E  and  mixes  with  the  steam  in  the  com- 
bining tube  C,  and  is  carried  forward  together  with 
the  condensed  steam  with  great  velocity  to  the 
delivery  tube  D,  thence  into  the  boiler.  The  maxi- 
mum velocity  is  reached  at  the  narrowest  part  of  the 
delivery  tube.  The  break  at  0  is  the  overflow  to 
allow  the  excess  of  water  or  steam  to  escape.  It  is, 
as  we  shall  see  in  the  next  chapter,  highly  desirable 
that  the  temperature  of  feed-water  should  be  raised 
to  the  highest  possible  degree  at  which  the  injector 
will  work  before  it  enters  the  boiler ;  and  if  this  can 
be  accomplished  by  means  of  exhaust  steam,  which 
would  otherwise  go  to  waste,  it  is  easily  apparent 
that  a  great  saving  must  of  necessity  be  effected. 

An  injector  depending  for  its  working  mainly  on 
exhaust  steam  was  introduced  some  years  ago  by 
Messrs  Davies  &  Metcalfe,  and  recently  they  have 
greatly  improved  the  apparatus.  Leading  off  from 
the  blast-pipe  of  the  locomotive  is  a  branch  pipe, 
by  means  of  which  steam  is  conveyed  to  a  grease 
separator,  where  the  exhaust  steam  is  freed  from 
any  oily  impurities  or  water  present.  The  steam 
then  passes  to  a  central  exhaust  nozzle  8  (Fig.  23), 
at  the  mouth  of  which  it  comes  into  contact  with 
the  feed-water  from  E.  Condensation  takes  place, 
and  a  high  velocity  is  thus  imparted  to  the  combined 
jet,  which  then  flows  forward  through  a  draught 


62 


THE  MODERN  LOCOMOTIVE 


[CH. 


tube.  At  the  end  of  this  it  meets  with  a  second 
supply  of  exhaust  steam,  which  imparts  a  further 
supply  of  energy  to  it,  and  the  combined  jet  enters 
the  combining  nozzle,  (7,  where  complete  condensa- 
tion takes  place,  and  its  velocity  is  still  further 
increased.  Then  it  passes  to  the  delivery  nozzle,  D, 
where  its  velocity  energy  is  transformed  into  pressure 
energy  and  so  to  the  boiler,  F.  The  exhaust  steam 


Fig.  23.     Exhaust  steam  injector. 

is  capable  of  thus  developing  a  pressure  of  120  Ibs., 
and  for  the  additional  pressure  required  to  force  the 
water  into  the  boiler  a  small  jet  of  live  steam  is 
introduced  through  a  supplementary  nozzle. 

Another  form  of  injector  using  live  steam  from 
the  boiler  has  warming  cocks  fitted,  so  as  to  enable 
the  driver  to  blow  surplus  boiler  steam  into  the 
water  tank  whenever  the  safety  valves  are  lifting ; 


iv]  SUPERHEATING,  ETC.  63 

most  drivers,  however,  prefer  to  keep  the  water  in 
their  tenders  cold,  especially  if  there  is  any  doubt 
as  to  the  ability  of  the  injector  to  deal  with  hot  water. 

No  doubt  hot  feed  will  become  more  extensively 
used  when  locomotive  superintendents  are  thoroughly 
convinced  of  the  modern  injector's  capability  to  pass 
hot  water  with  the  same  certainty  with  which  it 
takes  cold  water. 

The  chief  and  most  important  means  of  increasing 
the  efficiency  of  the  steam  is  by  superheating  which, 
together  with  the  methods  of  feed  heating  and 
thermal  storage,  will  claim  our  attention  in  the 
next  chapter. 

CHAPTER   IV 

SUPERHEATING,   THERMAL   STORAGE,   FEED 
HEATING 

THE  effect  of  heat  upon  water  is  to  convert  it  into 
steam.  That  portion  of  the  heat  which  produces  the 
necessary  rise  in  temperature  is  called  the  sensible 
heat.  Thus,  to  raise  the  temperature  of  one  pound 
of  water  from  32°  F.  to  212°  F.  or  through  180° 
requires  practically  180  British  Thermal  Units*. 

*  In  the  production  of  one  B.T.U.  it  is  usually  stated  that  772  ft.-lbs. 
of  mechanical  energy  disappear.  Later  investigations,  however,  give 
774  and  778.  The  original  figure  is  accurate  enough  for  all  ordinary 
investigations. 


64  THE  MODERN  LOCOMOTIVE          [OH. 

Or  h  =  t°F.-  32°, 

where  h  =  the  sensible  heat. 

After  having  reached  the  boiling  point  the  water 
gradually  disappears  until  the  whole  of  the  1  Ib.  of 
water  has  been  converted  into  1  Ib.  of  steam,  during 
which  process  the  temperature  remains  constant  at 
212°  F.  The  heat  thus  imparted  to  produce  the 
change  of  state,  without  change  of  temperature,  is 
called  latent  heat.  The  conversion  of  1  Ib.  of  water 
to  1  Ib.  of  steam  absorbs  967  B.T.TJ.  Note  however 
that  this  quantity  is  true  only  for  steam  formed  at 
the  pressure  of  one  atmosphere. 

The  latent  heat  may  be  approximately  obtained 
from  the  formula 

L  =  IU4-0'7t°~F. 

where  £  =  the  latent  heat  in  thermal  units  of  one 
pound  of  steam  formed  at  a  temperature  t°  F. 

Steam  in  contact  with  the  water  from  which  it  is 
generated  is  known  as  saturated  steam  and  is  steam 
at  its  maximum  density.  After  the  water  has  com- 
pletely disappeared,  if  heat  be  still  applied,  the 
temperature  as  before  will  rise,  provided  the  pressure 
is  maintained  constant :  it  is  then  known  as  super- 
heated steam. 

Saturated  steam  is  that  used  generally  in  loco- 
motive work  :  more  recently  superheated  steam  has 
been  used.  To  understand  the  properties  of  both 


IV] 


SUPERHEATING,  ETC. 


65 


a  knowledge  of  the  relation  between  pressure,  tem- 
perature, and  volume  is  essential.  These  relations 
have  been  obtained  by  Regnault  from  experimental 
data  and  the  values  met  with  in  locomotive  boiler 
working  are  given  in  round  numbers  in  the  following 
table. 

Properties  of  Saturated  Steam 


Gauge 
Pressure 
of  boiler 
(Ibs.  per 
sq.  inch) 

Temperature 
(°.F.) 

Total  Heat 
(Thermal 
Units) 

Latent  Heat 
(Thermal 
Units) 

Volume 
of  1  Ib. 
(in  cub. 
feet) 

150-3 

365-7 

1193-5 

855-1 

2-72 

160-3 

370-5 

1194-9 

851-6 

2-58 

170-3 

375-1 

1196-3 

848-2 

2-45 

180-3 

379-5 

1197-7 

845-0 

2-33 

190-3 

383-7 

1199-0 

841-9 

2-22 

200-3 

387-7 

1200-2 

838-9 

2-12 

215-3 

393-6 

1202-0 

835-8 

1-98 

225-3 

397-3 

1203-1 

833-1 

1-9 

It  may  be  stated  generally  that  the  pressure 
varies  with  the  temperature,  the  rate  of  change  of 
pressure  increasing  more  rapidly  as  the  temperature 
increases.  A  formula  expressing  the  connection 
between  the  temperature  and  pressure  of  saturated 
steam  given  by  Rankine  is  as  follows : 

6-1007-  ~ -~> 


A.  L. 


66  THE   MODERN  LOCOMOTIVE          [OH. 

in  which 

T=t  +  461°  F. 

(the  formula  for  converting  the  Fahrenheit  scale  to 
the  scale  of  absolute  temperature), 

log  B  =  3-4364, 

log  C  =5-5987. 

For  all  ordinary  purposes  in  connection  with 
locomotive  investigation,  however,  the  tables  suffice. 
The  connection  of  pressure  and  volume  is  usually 
expressed  by  the  formula  (also  by  Rankine)  : 


where 

P  =  pressure  in  Ibs.  per  sq.  inch, 

V—  volume  in  cubic  feet  per  pound  of  steam. 

The  total  heat  of  steam  is  the  total  of  the  sensible 
and  latent  heat  required  to  raise  the  temperature  of 
one  pound  of  water  from  32°  F.  and  convert  it  into 
saturated  steam  at  any  given  temperature.  Thus, 
according  to  definition, 


where 

H  is  the  total  heat, 
h  the  sensible  heat,  and 
L  the  latent  heat. 

But  we  have  seen  that  L  may  be  expressed 
1114-07S0,     and    A  =  Z°F.-32°, 


iv]  SUPERHEATING,  ETC.  67 

whence  we  get 

H  =  (t°  F.  -  32°)  +  (1 1 14  -  07  t°) 
=  1082  +  0-305  tQF. 

The  factors  H  and  L  are  given  in  the  table,  and  h 
may  be  obtained  by  subtracting  the  figure  in  column 
4  from  that  in  column  3. 

Now  locomotive  steam  is  generally  very  *  wet,' 
i.e.  it  contains  suspended  moisture,  due  to  the  violent 
ebullition  and  the  small  water  surface  available  for 
the  steam  to  escape  from.  The  dryness  fraction 
which  is  used  to  express  this  condition  of  the  steam 
averages  about  10  per  cent.  The  presence  of  this 
moisture  means  that  less  heat  is  required  than  is 
necessary  to  produce  the  same  weight  of  dry  steam, 
but  this  is  no  advantage  since  wet  steam  is  not  only 
very  undesirable  in  the  cylinders,  but  represents  coal 
burnt  to  no  purpose.  It  is  obvious  then  that  dry 
steam  would  mean  a  considerable  saving.  This  and 
more  is  obtained  by  using  superheated  steam. 

We  have  seen  that  superheated  steam  results  from 
a  continued  application  of  heat  to  the  steam  after  all 
the  water  has  been  evaporated.  What  happens  is 
that  its  temperature  then  becomes  more  than  that 
due  to  the  pressure,  a  state  impossible  with  saturated 
steam  which  has  only  one  temperature  for  a  given 
pressure.  If,  as  we  shall  see  happens  in  the  engine 
cylinders,  heat  is  abstracted  from  saturated  steam, 

5—2 


68  THE  MODERN  LOCOMOTIVE         [CH. 

its  temperature  is  not  lowered  but  some  of  it  is 
condensed  into  water.  On  the  other  hand  the  ad- 
dition of  heat  at  constant  pressure,  that  is  to  say, 
under  conditions  which  permit  the  steam  to  expand 
as  it  is  heated,  causes  a  rise  in  temperature.  Such 
steam  is  no  longer  saturated  but  superheated.  In 
this  state  and  in  proportion  to  its  temperature  rise  it 
behaves  less  like  a  vapour  and  more  like  a  perfect 
gas,  one  result  of  which  is  that  its  volume  per  pound 
also  increases  at  a  rate  roughly  proportional  to  the 
increase  of  its  absolute  temperature.  Its  temperature 
may  also  be  reduced  without  condensation.  Super- 
heated steam  has  a  greater  volume  than  the  same 
weight  of  saturated  steam,  the  increase  in  volume 
being  roughly  12J  per  cent,  for  every  100°  F.  of  super- 
heat added.  Its  specific  heat  does  not  appear  to  be 
constant,  but  for  practical  purposes  it  may  be  taken 
as  equal  to  0*48  at  constant  pressure. 

To  ascertain  the  total  heat  required  to  form 
superheated  steam,  the  total  heat  of  saturated  steam 
at  the  given  pressure  is  first  found  according  to  the 
equation  stated  above,  to  which  is  added  the  heat 
required  to  superheat  the  steam  given  by 

0-48  (*.-  O, 
where 

tg  =  the  temperature  due  to  superheating, 
ti  =  the  temperature  of  the  boiling-point  due 
to  the  pressure. 


iv]  SUPERHEATING,   ETC.  69 

It  is  evident  therefore  that  with  superheating 
additional  heat  is  required,  which,  however,  is  carried 
as  an  increased  number  of  units  per  pound  of  steam 
to  the  cylinder  with  a  very  considerable  effect  upon 
efficiency.  In  the  first  place  the  loss  occasioned  by 
wetness  carried  over  from  the  boiler  is  removed,  and 
that  due  to  initial  condensation  and  heat  interchange 
between  the  steam  and  cylinder  walls  is  reduced  to 
an  extent  dependent  upon  the  degree  to  which  super- 
heating is  carried.  The  last  mentioned  losses  need 
explanation.  They  are  due  to  the  action  of  the 
piston  in  a  cylinder.  As  it  moves  up  the  cylinder 
the  pressure  of  the  steam  is  reduced  by  expansion, 
consequently  the  temperature  is  reduced.  This  means 
that  condensation  takes  place  to  form  water.  The 
condensed  steam  is  partly  re-evaporated  by  the  next 
inrush  of  steam,  but  this  robs  it  of  its  heat,  and  so 
reduces  its  efficiency  of  work.  This  heat  exchange  is 
continually  going  on  at  every  stroke  of  the  piston, 
and,  in  fact,  the  formation  of  a  film  of  water  on  the 
metal  surface  of  the  cylinder  constitutes  the  heaviest 
loss  in  the  expansive  working  of  steam. 

As  superheated  steam  cannot  become  condensed 
until  the  temperature  has  fallen  back  to  its  saturation 
or  boiler  temperature,  it  becomes  more  stable,  and  it 
is  thus  possible  to  use  the  steam  in  the  cylinders  in 
a  dry  state  without  any  losses  due  to  liquefaction. 
A  higher  theoretical  efficiency  is  thus  obtained  from 


70  THE  MODERN  LOCOMOTIVE         [OH. 

the  steam  owing  to  its  greater  elasticity  ;  also,  as 
one  effect  of  superheating  is  to  increase  the  volume 
occupied  by  a  given  weight  of  steam  without  altering 
its  pressure,  a  less  weight  of  steam  is  required  per 
stroke.  Prof.  Ripper  states  that  7*5°  of  superheat 
are  sufficient  to  compensate  for  loss  due  to  1  per  cent, 
of  initial  condensation,  and  he  has  shewn  that  the 
heat  exchange  between  the  steam  and  cylinder  walls 
is  correspondingly  reduced. 

It  will  probably  occur  to  the  reader  that  this  is 
all  very  well,  but  only  a  given  quantity  of  heat  can 
be  generated  in  the  fire-box,  and  if  a  portion  of  this 
is  used  for  superheating,  so  much  the  less  is  available 
for  producing  steam ;  in  fact  it  appears  to  be  a 
question  of  taking  a  penny  out  of  one  pocket  and 
putting  it  into  another.  With  this  in  view  how 
exactly  is  the  increased  efficiency  to  be  accounted 
for?  No  one  has  explained  this  more  clearly  than 
Prof.  Ripper*.  Suppose  an  engine  using  saturated 
steam  with  25  per  cent,  of  the  steam  condensed  up  to 
the  point  of  cut-off  Then  since  1  per  cent,  of  wetness 
requires  7*5°  F.  of  superheat,  25  per  cent,  of  wetness 
will  require  7*5  x  25  =  187*5°  F.  of  superheat.  But 
the  specific  heat  of  superheated  steam  is  0*48,  hence 
there  is  required  187*5  x  0'48  =  90  thermal  units  per 
pound  of  steam.  As  there  is  only  75  per  cent,  of  the 
steam  engaged  in  doing  useful  work,  approximately 

*  The  Steam  Engine  in  Theory  and  Practice. 


iv]  SUPERHEATING,  ETC.  71 

1000  heat  units  per  pound  of  steam  would  be  supplied, 
and  putting  the  heat  efficiency  at  10  per  cent.  100 
out  of  the  1000  units  are  converted  into  work.  By 
supplying,  as  above,  90  thermal  units  as  superheat, 
the  whole  of  the  steam  present  in  the  cylinder  is 
'dry'  and  the  useful  work  done  is  increased  approxi- 
mately in  the  proportion  of  from  75  to  100  =  a  gain 
of  33  per  cent.  This  gives  133*3  heat  units  converted 
into  work  out  of  a  total  of  1090,  or  an  efficiency  of 
12*23  per  cent,  as  against  10  per  cent,  without  super- 
heat. The  portion  of  the  heat  used  for  superheating 
thus  shews  the  high  efficiency  of 

33*3 

— -  x  100  =  37  per  cent. 

So  much  for  the  theory  of  the  subject.  When 
applied  in  practice  we  should  expect  to  see  this 
increased  efficiency  represented  by  a  saving  in  coal 
and  water  for  a  given  power.  Numerous  tests  carried 
out  on  actual  locomotives  both  when  stationary  and 
on  the  road  shew  that  economy  in  fuel  and  water  and 
increased  efficiency  are  so  obtained  and  in  proportion 
to  the  increasing  degree  of  superheat.  To  cite  one 
only  of  numerous  elaborate  locomotive  tests,  namely, 
that  carried  out  by  Prof.  Goss  at  the  Purdue  Univer- 
sity, it  was  found  that  the  substitution  of  superheated 
for  saturated  steam  for  a  given  fixed  power  permits : — 

The  use  of  comparatively  low  steam  pressures,  a 


72  THE  MODERN  LOCOMOTIVE         [OH. 

generally  accepted  limit  being  160  Ibs. :  a  saving  of 
from  15  to  20  per  cent,  in  the  amount  of  water  used :  a 
saving  of  from  10  to  15  per  cent,  in  the  amount  of  coal 
used  while  running,  or  of  from  3  to  12  per  cent,  in  the 
total  fuel  supplied  :  assuming  the  power  developed 
to  equal  the  maximum  capacity  of  the  locomotive 
in  each  case,  the  substitution  of  superheated  for 
saturated  steam  will  permit  an  increase  of  from  10 
to  15  per  cent,  in  the  amount  of  power  developed, 
accompanied  by  a  reduction  in  total  water  consump- 
tion of  not  less  than  5  per  cent,  and  by  no  increase  in 
the  amount  of  fuel  consumed. 

In  tests  carried  out  in  actual  working  Mr  Hughes, 
of  the  Lancashire  and  Yorkshire  Railway,  shewed  that 
a  superheater  engine,  when  put  to  the  highest  test, 
that  is  by  running  against  a  compound  engine,  gave 
results  which  represent  an  economy  in  total  coal  per 
train-mile  of  12*6,  and  per  ton-mile  of  12*4  per  cent, 
in  favour  of  the  superheater. 

Again,  to  take  the  most  recent  results  available, 
Mr  Bierman,  of  the  Dutch  Railway  Company,  gives 
as  the  results  of  carefully  made  runs  with  express 
and  ordinary  trains  a  saving  of  2*17  kgs.  per  train- 
kilometre  (7'70  Ibs.  per  train-mile)  representing  for 
the  seven  months  which  the  locomotives  were  em- 
ployed on  trial,  a  saving  of  377,455  kgs.  (832,145  Ibs.) 
of  coal  in  running  173,972  locomotive-kilometres 
(108,103  locomotive-miles). 


iv]  SUPERHEATING,  ETC.  73 

The  impulse  was  first  given  to  the  now  widely 
prevailing  movement  of  superheating  by  its  re- 
introduction  in  1898  on  German  engines.  [It  is 
not  generally  known  that  as  far  back  as  1845,  a 
Gt  Western  engine  was  fitted  out  with  a  superheater 
and  that  in  the  early  fifties  MacConnell,  on  the 
London  and  North  Western  Railway,  also  employed 
superheaters  on  some  of  his  locomotives.  These  ex- 
periments were  apparently  in  advance  of  their  time.] 
Belgium,  France,  Switzerland  and  America  followed  to 
the  extent  that  the  practice  has,  in  combination  with 
compounding,  become  standard  in  these  countries. 
British  engineers  were  slower  to  convince,  but 
after  careful  and  tentative  trials,  the  practice  is 
steadily  advancing,  and  locomotives  so  fitted  are 
found  on  most  of  our  leading  railways. 

As  an  example  of  a  superheating  apparatus  em- 
ployed in  Great  Britain,  that  introduced  last  year  by 
Mr  C.  J.  Bowen  Cooke  on  the  London  and  North 
Western  Railway  is  illustrated  in  Fig.  24.  The  lower 
rows  of  tubes  A,  which  carry  the  furnace  gases  from 
the  fire-box  to  the  smoke-box,  are  of  the  usual  type 
and  diameter.  The  upper  rows  B  are  much  larger, 
and  in  these  larger  tubes,  twenty-four  in  number,  the 
steam  superheater  tubes  are  arranged.  When  the 
steam  regulator  valve  D  is  opened  the  steam  passes 
from  the  boiler  along  the  main  steam  pipe  E  to  a 
steam  collector  F  fixed  on  the  front  of  the  boiler. 


74 


THE  MODERN  LOCOMOTIVE         [CH. 


1 

«*H 

O 

OQ 


iv]  SUPERHEATING,  ETC.  75 

The  steam  collector  is  divided  into  compartments,  a 
saturated  steam  chamber  receiving  the  steam  direct 
from  the  boiler  at  a  temperature  of  about  377°  F., 
and  the  superheated  steam  chamber  receiving  the 
steam  from  the  superheater  tubes  at  a  temperature 
of  about  650°  F.,  passing  from  which  it  passes  to  the 
cylinders.  One  end  of  each  superheater  tube  is 
connected  to  the  saturated  steam  chamber,  whence  it 
runs  along  the  large  tube  B,  nearly  up  to  the  fire-box 
and  back  to  the  superheated  steam  chamber,  to 
which  the  other  end  is  connected.  The  steam  on  its 
way  from  the  boiler  to  the  cylinders  thus  passes 
through  the  regulator  valve  to  the  steam  collector, 
through  the  superheater  tubes  and  back  to  the 
steam  collector,  and  thence  by  the  steam  pipe  to  the 
cylinders.  It  becomes  superheated  to  a  maximum  of 
about  650°  F. 

The  temperature  of  the  superheated  steam  is 
measured  by  a  pyrometer  connected  to  the  super- 
heater chamber  of  the  steam  collector,  and  is  indicated 
by  a  gauge  fixed  in  the  engine  cab  under  the  obser- 
vation of  the  engine  driver.  In  order  to  regulate  the 
amount  of  superheat  a  movable  plate  H  is  fixed  on 
the  smoke-box  tube  plate,  or  front  of  the  boiler 
barrel,  by  means  of  which  the  temperature  of  the 
heated  gases  passing  through  the  large  fire  tubes 
may  be  controlled,  and  the  temperature  of  the  steam 
passing  though  the  steam  tubes  within  them  regulated. 


76  THE  MODERN  LOCOMOTIVE         [CH. 

Feed-Water  Heating.  The  method  almost  in- 
variably resorted  to  in  stationary  engine  practice  of 
utilizing  the  residuum  of  heat  in  the  exhaust  steam 
or  the  flue  gases  after  leaving  the  flues,  for  heating 
the  water  fed  into  the  boiler,  has  not  yet  found 
extensive  imitation  in  locomotive  design. 

This  is  probably  due  to  the  fact  that  until  recently 
it  was  not  possible  to  adapt  the  injectors  to  the  work 
of  feeding  hot  water,  and  that  English  locomotive 
practice  had  discarded  feed  pumps  in  favour  of 
injectors.  Feed  pumps,  successfully  to  replace  in- 
jectors, must  be  independent  steam-driven  units. 
These,  however,  take  up  room,  and  perhaps  necessitate 
more  attention  from  the  driver.  Nevertheless,  Mr 
Drummond  on  the  London  and  South  Western  Rail- 
way has  tackled  the  problem  vigorously  and  fitted  a 
number  of  his  engines  with  feed- water  heaters.  The 
apparatus  is  illustrated  in  Fig.  25.  The  exhaust  steam 
from  the  pumps  which  deliver  the  hot  feed-water  to 
the  boiler  is  sent  to  the  tender  with  that  portion  of 
the  main  exhaust  utilized  for  the  purpose.  The  water 
is  pumped  into  the  boiler  at  a  temperature  of  about 
180°  F.  Mr  Drummond  states  that  a  saving  in  fuel 
equal  to  6  Ibs.  per  mile  is  effected.  The  tank  from 
which  the  feed  is  immediately  drawn,  and  through 
which  the  exhaust  steam  heating  pipes  are  led,  is 
supplementary  to  the  tender  tank.  The  condensed 
water  escapes  through  a  series  of  holes  in  the  rear 


IV] 


SUPERHEATING,  ETC. 


77 


casting  which  receives  the  ends  of  the  pipes,  whilst 
the  uncondensed  steam  passes  into  the  atmosphere 
through  an  escape  pipe  at  the  rear  of  the  tender. 


Fig.  25.     Feed-water  heating  apparatus  ;  London  and  South 
Western  Eailway. 

A     Steam  cylinder  (5£  ins.  diam.      D    Delivery  from  pump. 

9  ins.  stroke). 
B     Pump  (4^  ins.  diam.  9  ins. 

stroke). 
C     Valve  box. 


Exhaust  steam  from  cylinder. 
Pump  suction. 


G    Exhaust  steam  from  pump. 


Thermal  Storage.  Mr  Druitt  Halpin's  system 
of  thermal  storage  as  applied  to  steam  boilers  for 
stationary  engines  has,  in  certain  circumstances, 


78  THE  MODERN  LOCOMOTIVE          [OH. 

shewn  itself  to  possess  distinct  advantages  over 
ordinary  methods  of  boiler  feeding.  A  test  conducted 
by  Prof.  Unwin  with  Cornish  boilers  shewed  a  coal 
saving  of  197  per  cent.  In  order  to  ascertain  the 
increased  efficiency,  if  any,  due  to  the  application  of 
the  system  to  locomotives,  Mr  Ivatt,  when  locomotive 
superintendent  of  the  Gt  Northern  Railway,  fitted 
a  2-4-0  type  passenger  engine  with  the  Halpin 
apparatus.  The  arrangement  is  very  simple,  and 
consists  of  a  cylindrical  storage  tank  placed  above 
and  connected  to  the  boiler  by  means  of  a  pipe.  All 
the  feed-water,  which  is  maintained  at  or  about  the 
same  temperature  as  the  water  in  the  boiler,  is  passed 
through  the  cylinder,  the  water  being  heated  by 
steam  generated  during  the  intermittent  periods  when 
the  engine  is  standing  or  the  safety  valves  are  blowing. 
The  water  thus  heated  is  fed  to  the  boiler  as  required 
when  the  engine  is  running,  this  being  regulated  by 
a  valve  in  the  driver's  cab.  Six  tank  engines  on  the 
Lancashire  and  Yorkshire  Railway  were  some  time 
ago  equipped  with  this  apparatus.  Where  stopping 
places  are  frequent  and  on  rising  gradients  Mr  Hughes 
states  that  there  is  distinct  economy.  Certain  tests 
were  carried  out  between  Salford  and  Accrington, 
resulting  in  an  actual  saving  of  1  ton  of  water,  and 
under  similar  conditions  elsewhere  the  saving  was  12 
per  cent.  When,  however,  these  engines  have  to  take 
their  turn  on  other  sections  of  the  line  which  are  not 


v]    RESISTANCE,  TRACTIVE  EFFORT,  ETC.    79 

so  favourable,  the  all-round  economy  is  brought  down 
to  4  per  cent. 


CHAPTER  V 

RESISTANCE,   TRACTIVE  EFFORT,   ADHESION 

HAVING  seen  how  the  steam  is  generated  the 
question  arises,  what  work  is  to  be  done  by  it  ?  The 
engine  has  not  only  to  propel  itself  but  to  overcome 
the  resistance  offered  by  the  train.  The  combined 
resistance  is  made  up  of  several  components. 
(1)  Resistance  dependent  on  the  speed.  (2)  The 
resistance  caused  by  flange  action  and  weather. 
(3)  Resistance  due  to  gradient.  (4)  Rolling  and  axle 
friction  and  side  play.  Resistance  dependent  on  the 
speed  is  due  to  the  friction  of  the  mechanism  of  the 
engine  and  the  air  resistance  due  to  engine  frontage. 
The  determination  of  this  is  still  a  subject  of  investi- 
gation, and  various  formulae  are  proposed  from  time 
to  time,  the  results  obtained  from  which,  however, 
do  not  appear  to  agree  very  closely  amongst  them- 
selves. Mr  Daniel  Gooch,  of  the  Great  Western 
Railway,  conducted  a  number  of  experiments  during 
the  gauge  controversy,  from  which  D.  H.  Clark 
obtained  the  much  used  formula 


80  THE  MODERN  LOCOMOTIVE          [CH. 

where 

V  =  the  velocity  in  miles  per  hour 
R  —  train  resistance  in  pounds  per  ton 

for  engine  and  vehicles  combined,  which  is  based  on 
the  assumption  that  the  rolling  stock  and  rails  are 
in  good  condition,  and  assuming  an  absence  of  side 
wind  and  wet.  Lubrication  at  that  period  (1855)  was 
effected  with  grease,  which  has  since  been  replaced 
by  oil,  thereby  reducing  axle  friction  from  about 
6  Ibs.  per  ton  to  between  3  and  4  Ibs.  at  slow  speed, 
the  resistance  rising  as  the  speed  increases.  To  meet 
this  the  formula  (1)  was  modified  to 

...............  (2). 


It  may  be  noticed  too  that  so  far  as  the  interaction 
between  wheel  and  rail  is  concerned,  rolling  friction 
has  been  reduced  by  the  adoption  of  steel  for  tyres 
and  rails. 

A  good  working  formula  proposed  by  Pettigrew  is 

(3). 


Later  still  M.  Barbier  of  the  Chemin  de  Fer  du  Nord 
presented  formulas  which  in  construction  have  been 
followed  by  nearly  all  other  investigations. 

The  following  table   gives  the  most  important 
results  in  tabulated  form. 


v]    RESISTANCE,  TRACTIVE  EFFORT,  ETC.    81 


Formulas  for   Train  Resistance 

E  =  Tractive  resistance  in  Ibs.  per  ton  (2240  Ibs.). 
V=  Speed,  miles  per  hour. 
L  =  Length  of  train  in  feet. 


No. 

Authority 

Formula 

Remarks 

1 

2 
3 

4 
5 
6 

7 
8 

Clark 

Sinclair 
Pettigrew 

Deeley 
Barbier 

J5 
J> 

A  spin  all 

V2 

8+fn 

2  +  0-24F 
9  +  0-007F2 

T72 

3  +  290 

58  ,  i-essr^1'609^60^ 

Whole 
train 

4  -wheeled 
Vehicles 

Bogie 
Vehicles 

Engine  and 
Tender 

Bogie 
Coaches 

3-                \     1000     ) 

3-59,l-611T-P609F+10^ 

x  (     1000     I 

e.51[3.o1F..(l-609F  +  30\ 

V     1000     ) 

V* 
O.K  ,                r 

1  50-8  +  0-0278L 

Resistance  due  to  Gradient.  In  addition  to  over- 
coming the  friction  of  the  mechanism,  the  engine 
must  be  able  to  haul  its  load  up  inclines.  The  effect 
of  gravity  against  ascending  an  incline  can  be  ex- 
pressed by, 

R=  TFXsinfl,  where 

R  =  the  resistance  in  Ibs.  per  ton  hauled. 

A.  L.  6 


82  THE  MODERN  LOCOMOTIVE         [OH. 

TF=the  load  and  sin  0=.    ^rtical  rise 

length  of  incline 

Thus  if  W=  1  ton  and  sin  6  =  FJF  then 

1  x  2240 

— ^,r-  =  /  4  IDS.  per  ton  due  to  gravity. 
oUU 

In  comparison  with  axle  friction  this  represents  a 
factor  which  does  not  admit  of  reduction. 

It  is  seldom  possible  to  ascertain  the  actual 
weight  of  a  train,  but  if  the  number  of  axles  be 
counted  and  5  tons  allowed  for  each,  a  very  fair  esti- 
mate of  the  weight  of  a  passenger  train  can  be  made. 

Resistance  due  to  Curves.  When  a  train  runs 
through  a  curve,  especially  if  it  be  a  reverse,  or  S- 
curve,  a  large  amount  of  resistance  is  set  up  by  the 
grinding  action  of  the  wheel  flanges  against  the 
rails,  the  collars  of  the  axle  journals  being  forced 
against  the  bearings,  thus  developing  end  friction. 
Curve  resistance  depends  upon  the  radius  of  the 
curve  and  the  length  of  the  rigid  wheel  base  of  the 
vehicles.  It  is  a  rather  uncertain  quantity  involving 
the  state  of  the  rails,  whether  dry  or  greasy,  and  the 
strength  and  action  of  the  wind.  A  formula  due  to 
Morrison  is 

WF(D  +  L) 
~W~  "' 

where  R  =  resistance ;  W  =  weight  of  vehicle ;  F  =  co- 
efficient of  friction  between  wheel  and  rail  varying 


v]    RESISTANCE,  TRACTIVE  EFFORT,  ETC.    83 

according  to  weather  from  O'l  to  0'27  ;  D  =  distance 
of  rail  between  treads  ;  and  L  =  length  of  rigid  wheel 
base. 

Much  has  been  done  in  recent  years  to  reduce 
curve  friction  by  the  provision  of  better  arrange- 
ments for  end  wear,  lubrication  and  short-based 
bogies.  Increased  resistance  and  wear  are  occasioned 
by  large  flange  play.  The  wind  has  a  great  effect  in 
increasing  train  resistance.  A  head  wind  virtually 
increases  the  velocity  with  which  the  train  travels 
against  the  air.  This  resistance  reaches  a  maximum 
when  the  wind  is  blowing  at  right  angles  to  the  train 
and  produces  the  side  effect  similar  to  that  on  a 
curve.  Carus  Wilson  states  that  the  resistance  of 
the  air  with  a  train  of  bogie  coaches  running  at 
60  miles  per  hour,  amounts  to  about  one  half  of  the 
total  tractive  effort  required  to  haul  the  train.  It 
is  claimed  by  some  that  a  large  reduction  can  be 
made  by  the  adoption  of  wedge-shape  '  wind  cutters,' 
familiar  on  Bavarian  locomotives,  to  the  extent 
of  10  per  cent,  of  the  total  tractive  effort  with  a 
passenger  train.  Against  this,  however,  must  be  set 
the  fact  that  when  the  engine  is  running  round  a 
curve,  or  is  exposed  to  a  side  wind,  the  air  pressure, 
so  far  from  being  reduced,  is  intensified. 

Resistance  to  the  progressive  movement  of  a 
train  may  be  determined,  when  uniform  speed  has 
been  attained,  by  calculating  the  total  force  exerted 

6—2 


84  THE  MODERN  LOCOMOTIVE          [OH. 

by  the  aid  of  indicator  diagrams ;  then  deducting 
the  drawbar  pull,  as  denoted  by  a  dynamometer,  we 
have  for  difference  the  total  resistance  of  the  loco- 
motive alone.  A  second  method  of  determination  is 
to  shut  off  steam  at  any  given  point  and  to  calculate 
the  operative  force  from  the  speed  variation.  In 
applying  this  method  the  engine  is  usually  allowed 
to  come  to  a  standstill  on  a  downhill  gradient,  and 
the  resistance  to  motion  is  equal  to  the  retarding 
force  plus  the  acceleration  due  to  the  gradient. 

The  question  of  resistance  to  locomotives  running 
at  high  speeds  is  of  a  complex  character,  for  in 
addition  to  the  commonly  recognized  forces  causing 
resistance  there  are  others  of  more  obscure  character 
which,  being  apparently  developed  within  the  machine, 
give  rise  to  what  is  called  l  internal  resistance.' 

It  is  known  that  the  size  of  the  wheels  and  the 
arrangement  of  mechanical  features  have  a  very 
important  effect  on  the  running  of  an  engine.  This 
point  has  been  well  illustrated  by  Mr  Ivatt,  when  he 
was  chief  at  Doncaster,  by  means  of  diagrams  taken 
from  Great  Northern  engines  shewing  the  relation 
between  horse-power  and  drawbar  pull. 

Indicator  diagrams  shew  the  power  developed  in 
the  cylinders,  but  not  the  proportions  of  the  total 
power  exerted  in  the  form  of  drawbar  pull,  because — 
and  particularly  at  high  speeds — much  of  the  cylin- 
der power  is  absorbed  in  overcoming  the  internal 


v]    RESISTANCE,  TRACTIVE  EFFORT,  ETC.    85 


resistance  of  the  engine  itself.  With  increase  of 
speed,  internal  resistance  increases  and  drawbar 
pull  diminishes,  until  a  point  is  reached  at  which 
the  engine  is  only  able  to  move  itself  and  exerts  no 
pull  at  all  on  the  drawbar.  This  will  be  more  fully 
realised  by  an  examination  of  the  following  figures 
given  by  Mr  Ivatt. 

Comparison  of  Drawbar  Pull  for  Two  Locomotives 
at  Different  Speeds 


Speed  in  Miles 
per  hour 

Drawbar  Pull  in  Tons 

Eight-coupled 
Goods  engine 

Single  -wheeled 
Express  engine 

10 
20 
30 
40 
50 
60 
70 
80 

7'6 
4-6 
2-0 
0-9 

(0-1)* 

(3-8)* 
3-0 
2-5 
2-1 
1-8 
1-3 
0-8 
0-4 

*  Computed. 

While  simply  illustrating  the  behaviour  of  two 
extreme  types  of  engine,  the  table  helps  to  shew 
the  advantage  to  be  derived  from  what  is  termed  a 
'free  running'  engine. 


86  THE  MODERN  LOCOMOTIVE          [OH. 

The  amount  of  power  absorbed  by  a  locomotive 
is  something  astonishing  to  the  uninitiated.  Accord- 
ing to  Mr  Sisterson  the  power  absorbed  in  running 
an  engine  weighing  from  80  to  90  tons,  together  with 
its  tender,  amounted  to  between  800  and  900  I.H.P., 
when  the  speed  of  about  70  miles  on  the  level  was 
attained.  This  represents  a  resistance  of  very  nearly 
60  Ibs.  per  ton  of  engine  and  tender.  Taking  another 
example,  based  on  the  running  of  the  Precursor,  a 
4-4-0  type  of  engine  designed  by  Mr  Whale  in  1905 
for  the  London  and  North  Western  Railway,  it  was 
found  that  during  a  run  between  Crewe  and  Rugby 
at  61  miles  an  hour,  the  drawbar  pull  was  2  tons, 
equivalent  to  about  730  horse-power  while  the  engine 
was  developing  1174  horse-power.  Here  we  have 
1174  —  730  =  444  I.H.P.  representing  resistance  of  the 
engine  alone.  It  would  be  interesting  to  know 
exactly  what  becomes  of  such  very  considerable 
amounts  of  power,  but  no  one  is  prepared  with  a 
precise  explanation. 

Inertia.  When  a  train  is  started  from  rest  an 
accelerating  force  is  required  to  put  the  mass  of 
the  train  in  motion  in  addition  to  the  force  required 
to  overcome  frictional  resistance.  This,  however,  is 
independent  of  the  uniform  rate  of  motion  considered 
above,  and  applies  more  particularly  to  suburban 
tank  engines. 

Adhesion.    Closely  connected  with  the  load  drawn 


v]    RESISTANCE,  TRACTIVE  EFFORT,  ETC.    87 

is  the  adhesion  between  the  driving  wheels  and  the 
rail,  that  is  to  say,  the  friction  between  them  avail- 
able to  resist  slipping.  If  the  adhesion  is  not  at 
least  equal  to  the  resistance  the  wheels  will  rotate 
and  slip  on  the  rail  without  advancing.  The  ad- 
hesion is  equal  to  the  weight  on  the  driving  wheels 
multiplied  by  a  coefficient  which  depends  upon  the 
condition  of  the  surface  of  the  rail.  This  may  vary 
between  J  in  dry  weather,  to  ^  in  wet  when  the 
rails  are  greasy.  It  is  sufficiently  accurate  to  take 
the  value  n>  =  J.  The  weight  on  the  driving  wheels 
depends  on  the  wheel  arrangement  adopted.  With 
the  single,  2-2-2  type,  only  the  weight  of  a  single 
pair  of  wheels  is  utilized,  and  as  the  strength  of  the 
rail  imposes  a  limit  of  20  tons,  this  represents  the 
limit  of  adhesion  of  the  single  engine.  In  the  4-4-0 
and  4-6-0  types  two-thirds  or  more  of  the  total 
weight  of  the  engine  is  available  for  adhesion,  and 
in  the  case  of  goods  engines  of  0-6-0  and  0-8-0 
types  the  whole  of  the  weight  is  so  utilized.  Thus, 
the  resistance  to  be  overcome  is  a  determining  factor 
of  the  wheel  base. 

Tractive  Effort.  To  overcome  the  total  resistance 
of  the  train  the  tractive  effort  produced  by  the  action 
of  steam  on  the  piston  by  which  propulsion  is  deter- 
mined must  at  least  equal  it.  Let  the  area  of  the 

piston  be  —r- ,  the  stroke  =  I  and  the  mean  effective 


88  THE  MODERN  LOCOMOTIVE          [CH. 

pressure  =p  ;  then  the  work  done  by  the  cylinder 
will  be  p  --  ,  and  for  one  revolution  of  the  wheel, 

in  a  two-cylinder  engine,  p-jrdH.  Let  E  =  the  mean 
effort  necessary  to  propel  the  engine  and  train  ;  and 
the  distance  travelled  during  one  revolution  of  the 
wheel  TrD,  D  being  the  diameter  of  the  driving  wheel, 
the  work  done  is  then  irDE.  Equating  these  two 
values  we  get 

-rrDE  = 


T-,     dHp 

whence  E  =  —  ~-  . 

This  value  E  represents  the  mean  tractive  effort 
of  the  locomotive  ;  the  mean  pressure  p  is  only  a 
fraction  of  the  boiler  pressure  and  must  be  evaluated. 

Thus  for  an  engine  with  cylinders  18  ins.  in 
diameter  by  24  in.  stroke  6  ft.  driving  wheels,  and 
taking  the  mean  effective  pressure  at  80  per  cent,  of 
the  200  Ibs.  the  boiler  pressure 

18  x  18  x  24  x  -8  x  200 
Tractive  force  =  —  —  -=-— 

7  * 

=  25,920  Ibs. 

In  an  engine  working  compound  (see  chapter  on 
Compounding)  the  tractive  effort  is  thus  determined. 
Let  p  and  p±  be  the  mean  effective  pressures  in  the 
high-  and  low-pressure  cylinders  respectively,  d  and 
dl  the  respective  diameters,  I  the  stroke  common  to 


v]     RESISTANCE,  TRACTIVE  EFFORT,  ETC.    89 

both.     In  a  two-cylinder  compound  engine  the  work 
per  revolution  is 


and  the  tractive  effort 


In  a  four-cylinder  engine,  the  factors  d2  and  d? 
must  be  replaced  by  2d2  and  2c?!2  since  there  are  two 
high-  and  two  low-pressure  cylinders. 

Therefore  we  obtain 


The  above  formula  involves  the  determination  of 
the  mean  effective  pressure. 

Yon  Borries  has  given  the  following  rule  for  a  two- 
cylinder  compound  (the  result  must  be  multiplied  by 
2  for  a  four-cylinder  engine)  : 

4T_xZ> 
= 


where      d  =  Diameter  of  the  low-pressure  cylinder, 

T=  Tractive  effort, 

D  =  Diameter  of  driving  wheel, 

p  —  Boiler  pressure, 

$=  Stroke  of  piston, 

...     d2  x  p  x  s  /rtX 

whence  ^     ~  .................. 


90  THE  MODERN  LOCOMOTIVE         [OH. 

The  formula  used  by  Baldwin  for  estimating  the 
tractive  power  of  four-cylinder  compounds,  is  as 
follows  : 

(72x£xjP     exSxip 
D  D 

in  which  C=  Diameter  of  H.P.  cylinder  in  ins., 
c  =  Diameter  of  L.  P.  cylinder  in  ins., 

S  =  Stroke  in  ins., 

P  =  Boiler  pressure  in  Ibs., 

T  =  Tractive  power, 

D  =  Diameter  of  driving  wheel  in  ins. 
Another  formula  is 


whence  r  is  the  ratio  of  the  cylinder  volumes,  the 
other  equivalents  being  as  in  (2). 

Mean  Effective  Pressure.  In  the  locomotive  as 
indeed  in  all  steam  engines,  the  steam  is  used 
expansively.  Steam  is  admitted  during  the  period 
the  piston  is  performing  a  portion  of  its  stroke,  and 
the  valve  then  closes,  cutting  off  the  steam.  The 
steam  in  the  cylinder  then  expands,  expansion  con- 
tinuing and  the  pressure  diminishing  until  the  piston 
has  nearly  completed  its  stroke  when  the  exhaust 
takes  place,  and  the  pressure  falls  very  nearly  to  that 
of  the  atmosphere. 


v]    RESISTANCE,  TRACTIVE  EFFORT,  ETC.    91 

During  admission  the  pressure  is  practically 
uniform,  and  from  the  point  of  cut-off  until  the 
exhaust  commences  expansion  follows  very  closely 
Boyle's  law  :  pv  =  a  constant. 

In  the  locomotive  the  point  of  cut-oif  is  arranged 
to  take  place  from  75  per  cent,  of  the  stroke  down 
to  20  per  cent.,  according  to  the  nature  of  the  work 
required.  The  average  or  mean  effective  pressure 
on  the  piston  can  be  determined  either  from  an 
indicator  diagram  or  by  calculation. 

Readers  are  referred  to  a  text-book  on  the  steam 
engine  for  an  explanation  of  the  indicator  and  its 
method  of  use.  It  will  suffice  to  state  here  that,  by 
this  apparatus,  a  figure  or  diagram  is  traced  on  a 
piece  of  paper  representing  the  pressure  of  the  steam 
in  the  cylinder ;  the  upper  line  shews  the  pressure 
urging  the  piston  forward  and  the  lower  line  the 
pressure  retarding  its  movement  on  the  return  stroke. 
The  mean  effective  pressure  may  be  obtained  by 
calculation  from  the  equation 


r 
where 

Pm  =  Mean  effective  pressure. 


boiler  pressure,  plus  that  due  to  the 
atmosphere  =  15  Ibs. 


92  THE  MODERN  LOCOMOTIVE         [OH. 

jt?2  =  The  back  pressure  plus  that  due  to  the 
atmosphere  =  say  19  Ibs. 

r  =  The  ratio  of  expansion  calculated  by  divid- 
ing the  volume  of  steam  in  the  cylinder 
at  the  end  of  the  stroke  by  the  volume 
of  steam  in  the  cylinder  at  the  point  of 
cut-off,  i.e.  by  dividing  the  length  of 
stroke  by  the  cut-off.  It  may  be  put 
at  1*33  for  75  per  cent.,  and  5  for  20  per 
cent,  of  cut-off  respectively. 

loge  r  =  the  hyperbolic  logarithm  of  r,  the  ratio  of 
expansion.  For  r  =  1*33  and  r  =  5  the 
hyperbolic  logarithms  are  0*285  and  1*609 
respectively. 

For  example.     Let  the  pressure  be  175  Ibs.  =  190  Ibs. 
absolute.     Then  for  75  per  cent,  cut-off 


Pm  =  190  -  19  =  164-5  Ibs. 

J.  O«~> 

For  20  per  cent,  cut-off 

+  1-609)  . 


The  following  table  gives  a  few  hyperbolic  loga 
rithms  required  in  locomotive  practice. 


vi]          UTILIZATION  OF  THE  STEAM 

Hyperbolic  Logarithms 


93 


Eatio  of  Expansion 

Hyperbolic  Logarithms 

1-35 

0-3001 

2-0 

0-6931 

2-5 

0-9168 

3-0 

1-0986 

3-5 

1-2528 

4-0 

1-3863 

4-5 

1-5041 

5-0 

1-6094 

6-0 

1-7918 

7-0 

1-9459 

CHAPTER  VI 

UTILIZATION  OF  THE  STEAM 

THE  conversion  of  the  energy  of  the  steam  into 
the  work  necessary  to  overcome  resistance  and  thus 
propel  the  engine  itself  and  its  load  is  accomplished 
in  the  cylinders.  The  cylinder  is  a  cast-iron  casting 
the  interior  of  which  is  truly  bored  out  to  cylindrical 
shape,  to  afford  a  smooth  surface  for  the  recipro- 
cating motion  of  the  piston.  To  render  the  piston 
steam-tight,  grooves  are  turned  in  its  edge  into  which 
are  sprung  elastic  rings  made  of  steel  which  tend  to 
press  outwards  against  the  cylinder  walls.  A  piston 


94  THE  MODERN  LOCOMOTIVE         [OH. 

rod  is  attached  to  the  piston  by  means  of  a  nut  fitting 
a  screw  on  the  rod.  The  end  of  the  rod  is  tapered 
off  to  pass  through  a  tapered  hole  in  the  piston  which 
thus  prevents  it  becoming  slack  on  the  rod.  The  rod 
passes  through  a  hole  in  the  front  cylinder  cover,  the 
joint  being  made  steam-tight  by  means  of  a  stuffing 
box  containing  metallic  packing.  The  reciprocating 
motion  of  the  piston  and  its  rod  is  converted  in  a 
rotary  motion  at  the  crank  axle,  the  necessary  con- 
nection being  made  by  the  connecting  rod.  As  there 
are  usually  two  cylinders,  there  are  thus  two  cranks. 
These  are  set  at  an  angle  of  90°  to  each  other,  so  that 
when  one  piston  is  at  the  end  of  its  stroke,  or  on  the 
dead  centre,  the  other  is  in  its  position  of  maximum 
effort.  Steam  is  admitted  alternately  on  opposite 
sides  of  the  piston  through  two  steam  ports,  one  at 
each  end  of  the  cylinder,  leading  from  the  steam 
chest.  A  third  port,  called  the  exhaust  port,  allows 
the  steam  to  escape  to  the  blast-pipe.  These  ports 
open  into  the  steam  chest  in  which  the  slide  valve 
reciprocates  and  so  distributes  the  supply  of  steam  to 
the  ports  and  thus  to  the  piston.  The  valves  are 
driven  by  a  valve  gear  or  motion  driven  by  eccentrics 
on  the  main  shaft,  or  by  other  means  which  we  shall 
examine  later.  The  cylinders  are  arranged  at  the 
front  of  the  engine  generally  under  the  smoke-box 
and  either  inside  or  outside  the  frames.  The  pre- 
vailing British  practice  is  to  place  them  between  the 


vi]          UTILIZATION  OF  THE   STEAM  95 

frames,  which  method  imparts  a  rigidity  to  the  whole 
structure  since  the  cylinder  casting  itself  serves  as 
a  frame  stay.  Further,  the  effort  set  up  by  the  steam 
and  moving  parts  acting  at  a  minimum  distance 
from  the  longitudinal  axis  of  the  engine,  a  greater 
steadiness  in  running  is  obtained.  Foreign  practice 
generally,  however,  favours  the  outside  cylinder 
arrangement  in  that  it  permits  the  use  of  larger 
diameter  cylinders,  ready  accessibility  of  the  parts 
and  the  elimination  of  the  cranked  axle. 

In  compound  engines  three  and  four  cylinders 
are  employed,  the  high-pressure  cylinders  being 
arranged  outside  and  the  low-pressure  inside  the 
frames.  In  the  latest  engines  the  cylinders  are 
stepped,  that  is,  one  pair  is  set  in  advance  of  the 
other. 

The  steam  chests  occupy  a  position  corresponding 
to  the  type  of  valve  gear  employed.  With  interior 
cylinders  they  are  placed  either  between,  above,  or 
below  the  cylinders ;  with  outside  cylinders  the  gear 
is  also  generally  outside  and  the  steam  chests  placed 
on  top  of  the  cylinders,  sometimes  horizontally  and 
sometimes  inclined  towards  the  exterior.  The  out- 
side cylinder  engines  of  this  country  have  the  valve 
gear  and  steam  chests  disposed  inside  the  frames: 
in  American  engines  the  steam  chest  is  placed  out- 
side, above  the  cylinder,  communication  between 
the  valve  rod  and  valve  gear  being  made  through  a 


96  THE   MODERN  LOCOMOTIVE          [OH. 

rocking  shaft.  Cylinders  are  always  made  from  a 
hard  close-grained  cast  iron  and  when  of  the  inside 
type,  are  generally  cast  in  pairs.  Quite  recently  the 
method  of  casting  them  en  bloc  has  been  adopted 
thus  doing  away  with  a  joint  and  increasing  the 
rigidity. 

Pistons  are  usually  of  cast  steel  with  cast  iron  or 
cast  steel  piston  rings,  which,  when  in  position,  are 
about  3^  in.  open. 

Slide  Valves.  Valves  are  either  of  the  flat,  or 
'D'  type,  or  cylindrical  in  shape  when  they  are 
known  as  piston  valves.  Various  modifications  of 
the  old  flat  valve  have  been  introduced  in  recent 
years  with  the  object  of  reducing  the  excessive 
friction  between  the  valve  and  valve  face.  With 
these  the  steam  exerted  its  full  pressure  on  the 
whole  area  of  the  valve  back,  with  the  result  that 
a  large  percentage  of  the  power  developed  in  the 
cylinder  was  required  to  move  it.  With  flat  valves, 
what  is  called  'balancing'  is  now  largely  resorted  to, 
one  of  the  latest  designs  of  a  valve  so  modified  being 
shewn  in  Fig.  26. 

The  main  valve  consists  of  three  principal  parts ; 
the  valve  proper  AE,  the  balance  plate,  S,  and  the 
pressure  plate  above  it. 

The  valve  has  two  faces,  one  operating  on  the 
valve  seat  on  the  cylinder,  and  the  other  against  the 
face  of  the  balance  plate.  Both  faces  are  the  same, 


VI] 


UTILIZATION  OF  THE  STEAM 


97 


and  that  of  the  balance  plate  against  which  the  valve 
operates  is  a  duplicate  of  the  cylinder  valve  seat. 
The  walls  of  the  valve  are  provided  with  ports  AJE, 
which  pass  from  face  to  face  of  the  valve.  On  the 
opening  of  a  steam  port  the  pressure  has  free  access 
to  both  sides  of  the  valve  by  reason  of  the  passages 
AE  through  the  valve  to  the  port  F  in  the  face  of 
the  balance  plate,  which  corresponds  with  the  cylinder 
port.  Consequently  the  pressure  in  the  port  has  no 


Fig.  26.     Balanced  slide-valve. 

effect  upon  the  valve  as  it  acts  on  both  sides  of  the 
valve  face  in  equal  area  and  pressure. 

The  piston  valve  was  also  introduced  to  reduce  the 
frictional  resistance  to  the  valve  movement.  Briefly 
it  consists  (Figs.  10  and  27)  of  a  hollow  cylinder  turned 
at  the  ends  to  fit  a  bushing  in  which  the  steam  ports 
are  cut.  It  is  reduced  in  section  in  its  central  portion. 
The  ends  are  fitted  with  L-shaped  packing  rings, 
similar  in  construction  to  the  piston  rings  of  a 
cylinder,  and  uncover  the  steam  ports  to  steam  and 

A.  L.  7 


98 


THE  MODERN  LOCOMOTIVE 


[CH. 


exhaust  at  the  proper  time.  The  motion  of  the  valve 
and  steam  distribution  are  the  same  as  in  the  D 
valve,  but  as  the  pressure  does  not  act  in  forcing  it 
up  against  the  walls  of  its  bushing  or  seat,  it  is  easily 
driven.  There  is  just  enough  area  of  ring  to  make  it 
steam  tight  without  unnecessary  friction.  The  body 
is  usually  made  hollow  of  cast  iron  :  in  the  latest 
practice  seamless  steel  tubing  is  employed  with  light 
cast  steel  ends  riveted  on. 

-;*— JT*. 


Fig.  27.     Piston  valve  ;  Lancashire  and  Yorkshire  Kailway. 

The  greatest  disadvantage  under  which  the  piston 
valve  labours  is  its  inability  to  relieve  excess  pressure 
in  the  cylinder  port  by  lifting,  after  the  manner  of 
the  slide  valve.  This  renders  the  employment  of  a 
cylinder  relief  valve  imperative.  To  eliminate  the 
disadvantages  attaching  to  the  use  of  such  valves, 
Mr  Hughes  on  the  Lancashire  and  Yorkshire  Railway 


VI] 


UTILIZATION  OF  THE  STEAM 


99 


has  adopted  auxiliary  valves  on  the  piston  valve  itself. 
These  are  held  on  their  seats  as  long  as  the  pressure 
in  the  cylinder  does  not  rise  beyond  the  working 
pressure.  Should,  however,  an  excess  of  pressure 
arise  in  the  cylinder,  it  displaces  these  valves  and 
the  steam  is  discharged. 

Valves  of  the  lifting  or  i  poppet'  type  have  now  a 
considerable  vogue  on  Continental  locomotives.     One 


Fig.  28.     Poppet  valve  gear. 

such,  the  Lentz,  the  details  of  which  are  shewn  in 
Fig  28,  was  designed  more  particularly  to  meet  the 
conditions  set  up  by  the  use  of  highly  superheated 
steam  and  to  get  rid  of  a  complex  mechanism.  Each 
valve  is  screwed  on  to  a  steel  spindle  which  moves  up 
and  down  in  a  cast-iron  guide  and  is  rendered  steam- 
tight  by  means  of  turned  grooves,  G,  thus  rendering 
the  use  of  stuffing-boxes  unnecessary.  The  spindles 

7—2 


100  THE  MODERN  LOCOMOTIVE          [CH. 

end  in  broad  cylinder  heads  in  which  rollers,  K,  are 
arranged  in  such  a  manner  as  to  turn  easily.  The 
spindle  heads  slide  up  and  down  in  their  guides  with 
the  valves  and  isolate  the  upper  part  of  the  case  from 
the  lower  chamber.  The  upper  ends  of  the  spindle 
heads  carry  springs  for  loading  the  valves  to  ensure 
the  positive  closing  of  the  cam-rod  by  the  rollers. 
The  cam-rod,  T,  is  coupled  to  the  valve  gear  instead 
of  the  slide  valve  rod,  and,  in  moving  to  and  fro, 
opens  and  closes  the  four  different  inlet  and  outlet 
valves  by  means  of  the  cams  on  which  the  rollers, 
turning  freely  in  the  spindle  head,  move. 

With  the  object  of  reducing  the  waste  of  heat 
that  occurs  in  ordinary  locomotives  with  the  reversal 
of  direction  in  the  flow  of  steam,  the  Stumpf  valve 
gear  has  been  introduced.  With  this  apparatus 
the  steam  flows  in  a  continuous  direction  through 
the  cylinder,  the  inlet  valves  being  arranged  at  the 
ends,  and  does  not  cool  the  walls  except  at  the  centre 
of  its  length,  which  is  the  part  at  which  the  exhaust 
steam  escapes. 

The  admission  valves  are  of  the  double-seated 
type  with  springs.  Steam  is  admitted  almost  directly 
against  one  of  the  faces  of  the  piston,  and,  during 
the  latter  portion  of  the  piston  travel,  it  escapes 
through  ports  at  the  centre  into  the  exhaust  pipe. 
The  opening  of  the  valve  is  insured  by  a  mechanism 
similar  to  that  used  with  the  Lentz  valve. 


vi]          UTILIZATION   OF  TttK  StMAft     ;    18$ 

An  interesting  type  of  gear  in  use  on  the  Chicago 
and  North  Western  Railway  is  of  the  rotary  type 
and  represents  an  adoption  of  the  Corliss  principle 
to  suit  the  requirements  of  locomotive  practice. 
Two  valves  are  fitted  to  each  cylinder,  operating 
alternately  as  inlet  and  outlet,  and  driven  by  Corliss 
wrist  motion.  The  valve  gear  makes  use  of  the 
Stephen  son  link,  eccentrics  and  rocker  arm  as  far 
as  the  end  of  the  valve  stem.  This  is  connected 
to  a  wrist  plate,  which  has  hinged  attachments  to  a 
crank  arm  on  the  rotary  valve  spindle.  A  horizontal 
shaft  extends  across  the  back  of  the  cylinder  saddle, 
and  this  shaft  is  fitted  with  two  cranks  which  con- 
nect with  the  bearings  of  the  wrist  plates.  From 
the  centre  of  the  shaft  a  long  connecting  rod  extends 
back  to  a  short  crank,  so  that  when  the  link  is  raised 
or  lowered  from  the  central  position,  the  wrist  plate 
is  raised  to  regulate  the  lead.  The  valve  body  is 
journaled  in  the  heads  of  the  steam  chests,  and  its 
weight  is  supported  entirely  clear  of  the  valve  seat. 
The  valve  friction  is  thus  materially  reduced.  The 
valve  spindles,  in  their  passage  through  the  end  of 
the  steam  chests,  have  a  shoulder  which  forms  a 
steam-tight  bearing  and  requires  no  packing.  This 
type  of  valve  besides  being  efficient  is  stated  to  cause 
very  little  wear  on  the  machinery. 

Valve  Gear.  Of  scarcely  less  importance  than  the 
boiler  in  determining  the  efficiency  of  the  locomotive 


102  THE  MODERN  LOCOMOTIVE         [CH. 

as  a  whole  is  the  valve  motion,  upon  which  depends 
the  proper  distribution  of  the  steam.  The  system 
most  commonly  employed  is  the  gear  known  as  the 
Stephenson,  although  it  was  in  reality  invented 
by  William  Howe.  Others  extensively  used  and 
closely  resembling  it  are  the  Gooch  and  Allan.  The 
most  modern  systems  are,  however,  the  Walschaerts 
and  Joy  motions.  The  Stephenson  link  gear  is  also 
known  as  the  shifting  link  motion;  the  Gooch  is 
directly  opposite  in  its  action,  in  that  the  link  is 
stationary,  and  the  link  block  attached  to  the  valve 
rod  is  moved  up  and  down.  The  Allan  is  a  combina- 
tion of  the  Stephenson  and  Gooch  in  that  both  the 
link  itself  and  the  valve  rod  are  shifted.  All  of  these 
motions  are  operated  with  two  eccentrics,  one  for  the 
forward  and  the  other  for  the  backward  motion. 

In  the  Stephenson  and  Allan  motion  when  the 
eccentric  rods  are  open,  the  lead*  is  increased  as  the 

*  For  a  detailed  explanation  of  the  terms  'lead,'  'lap,'  the 
reader  is  referred  to  a  text-book  on  the  steam  engine.  It  may  'be 
stated  here,  however,  that  lead  is  the  amount  of  opening  of  the 
steam  port  at  the  beginning  of  the  stroke  of  the  piston.  Lap  is  the 
cover  of  the  steam  ports  by  the  outside,  or  steam,  edges  of  the  valve 
when  the  latter  is  at  its  mid  travel.  It  represents  the  distance  which 
the  valve  has  to  move  from  its  middle  position  to  open  either  steam 
port.  The  function  of  lap  is  to  give  a  varying  point  of  cut-off  and  so 
take  advantage  of  the  expansive  quality  of  the  steam.  The  simplest 
form  of  valve  gear  is  the  eccentric  and  its  rod.  The  eccentric  is  in 
reality  a  crank  whose  pin  is  so  enlarged  as  to  envelope  the  shaft. 
Its  eccentricity  or  '  throw '  is  the  distance  separating  the  centres  of 


VI] 


UTILIZATION  OF  THE   STEAM 


103 


link  is  pulled  up  and  the  point  of  cut-off  made  earlier. 
If,  however,  the  rods  are  crossed  the  'notching  up' 
reduces  the  lead,  though  this  reduction  is  much  less 
than  the  increase  in  the  former  case.  Again  with 
either  open  or  crossed  rods,  the  corresponding  in- 
crease or  reduction  of  lead  is  much  less  with  the 
Allan  than  with  the  Stephenson  valve  motion.  With 
the  Gooch,  Walschaerts,  and  Joy  motions  the  lead  is 
constant  for  all  points  of  cut-off. 

the  eccentric  and  the  shaft.  As  the  crank  shaft  rotates  the  valve  is 
driven  by  the  eccentric  to  and  fro  for  a  distance  equal  to  twice  the 
throw.  When  the  piston  is  at  the  end  of  its  stroke,  the  valve  will  be 
at  half  stroke  just  opening  to  steam  and  the  eccentric  is  placed  at  90° 
to  the  crank  (see  Fig.  29).  The  eccentric  leads  the  crank,  otherwise 
it  would  be  closing  the  steam  port  when  it  should  be  opening  it  to 
steam.  Taking  now  a  valve  in  its  mid  position  with  an  outside  lap 


Fig.  29.     Position  of  a  valve  without  lap  or  lead  and  of  a  valve 
with  lap  and  lead  at  the  beginning  of  the  stroke. 

L  (the  portion  shewn  in  black  in  Fig.  29).  To  uncover  the  steam 
port  it  must  be  moved  over  a  distance  equal  to  L,  and  the  crank, 
being  on  the  dead  centre,  the  eccentric  must  lead  the  crank  by  90°  +  lap 
or  distance  L.  But  we  have  seen  that  when  a  valve  has  lead,  the  steam 
port  is  already  open  at  the  beginning  of  the  stroke.  Thus  the  eccentric 
must  lead  the  crank  by  an  amount  equal  to  90°  + lap  and  lead. 


104 


THE   MODERN   LOCOMOTIVE 


[CH. 


The  Walschaerts  gear  is  driven  by  a  combination 
of  an  eccentric  or  short  stroke  return  crank  from  the 
main  crank-pin  and  a  connection  to  the  crosshead. 
The  Joy  gear  is  driven  from  a  connection  to  the 
connecting  rod.  The  former  has  been  extensively 
applied  on  the  Continent  of  Europe  and  the  latter 
is  in  use  on  the  London  and  North  Western,  and 
Lancashire  and  Yorkshire  railways. 


Fig.  30.     Diagram  of  Stephenson  link  motion. 

As  the  Stephenson  motion  is  the  one  mostly  used 
in  this  country  it  will  be  first  considered.  It  is 
illustrated  in  Fig.  30.  E,  E  are  the  two  eccentrics 
connected  to  a  slotted  link  L  at  the  points  P  and  P'. 
The  link  is  curved,  the  radius  of  curvature  being  that 
of  the  eccentric  rod.  It  is  capable  of  being  raised  or 
lowered  by  the  lever  K  and  accommodates  in  its 
slotted  portion  a  block  B,  which  slides  in  the  slot.  It 
is  directly  connected  to  the  valve  by  the  rod  V.  In 
the  position  shewn  on  the  diagram  the  block,  and 


vi]          UTILIZATION  OF  THE  STEAM         105 

therefore  the  valve,  is  not  influenced  by  the  motion  of 
either  eccentric  and  consequently  the  valve  theoreti- 
cally should  not  move.  Actually  however  it  moves 
very  slightly.  As  the  block  occupies  the  top  or 
bottom  position  of  the  link  it  is  brought  under  the 
influence  of  either  eccentric  E'  or  E  respectively  and 
the  valve  travels  its  full  distance.  At  intermediate 
positions  the  action  of  one  eccentric  is  more  or  less 
neutralized  by  that  of  the  other,  consequently  the 
travel  of  the  valve  is  reduced.  The  extreme  positions 
determine  whether  the  engine  will  run  in  a  backward 
or  forward  direction,  the  intermediate  positions  affect 
the  distribution  of  the  steam  and  the  rate  of  expan- 
sion by  altering  the  position  of  the  cut-oif.  The  lead 
and  point  of  release  of  the  exhaust  steam  are  also 
thus  capable  of  being  varied. 

The  motion  of  a  valve  driven  by  link  motion  is 
thus  exceedingly  complex;  it  can,  however,  be  ap- 
proximately determined  by  geometrical  methods,  the 
explanation  of  which  would  be  quite  outside  the 
scope  of  this  book.  The  lever  K  is  connected  by  a 
system  of  levers  to  a  quick  threaded  screw,  called 
the  reversing  screw,  on  the  foot  plate  by  which  the 
reversal  or  cut-off  is  obtained. 

Owing  to  the  considerable  manual  effort  required 
to  operate  the  reversing  gear  of  a  modern  locomo- 
tive, power  reversing  gear  has  been  applied  in  many 
cases.  One  system,  first  introduced  by  Mr  James 


106 


THE  MODERN  LOCOMOTIVE 


[CH. 


Stirling  on  the  Glasgow  and  South  Western  Railway, 
employs  a  steam  cylinder  controlled  and  locked  by 
a  hydraulic  cylinder  or  cataract  gear.  An  example 
of  that  in  use  on  the  Great  Eastern  Railway  worked 
by  compressed  air  is  illustrated  in  Fig.  31.  The 
locomotive  can  be  reversed  in  the  usual  way  by 
means  of  the  hand- wheel  A,  or  by  compressed  air, 


OPERATING  4  LOCKING  GEAR 


Fig.  31.     Power  reversing  gear  ;  Great  Eastern  Bailway. 

the  gear  for  this  purpose  being  operated  by  the 
handle  B.  The  reversing  shaft  C  is  connected  to  the 
reversing  rod  D  by  the  arm  E  and  to  the  piston  rod 
F  by  the  arm  G.  To  operate  the  gear  by  power,  the 
handle  B  is  moved  one  way  or  the  other,  according 
to  whether  the  engine  is  required  to  be  put  in 


VI] 


UTILIZATION  OF  THE  STEAM 


107 


forward  or  backward  running.  This  causes  the  valve 
L  to  rotate,  thereby  opening  one  end  of  the  reversing 
cylinder  K  to  pressure  and  the  other  end  to  exhaust. 
In  the  running  position  both  ends  of  the  cylinder  K 
are  in  communication  with  the  main  air  reservoir  P 
through  the  valve  L,  the  piston  rod  F  being  made  of 
such  a  diameter  that  the  reduced  piston  area,  ex- 
posed to  pressure  on  the  piston  rod  side  of  the 
piston,  balances  the  weight  of  the  motion  hanging  on 
to  the  lifting  links  Q.  Air  is  supplied  from  the 
Westinghouse  brake  pump. 


Fig.  32.     Joy's  valve  gear. 

The  Joy  valve  gear  is  an  example  of  what  is 
known  as  the  radial  valve  gear  and  is  almost  ex- 
clusively employed  on  the  engines  of  the  London  and 
North  Western  and  Lancashire  and  Yorkshire  rail- 
ways. Some  important  advantages  are  obtained 
with  this  gear,  the  chief  of  which  is  that  the  lead  is 


108  THE  MODERN  LOCOMOTIVE          [CH. 

constant  for  backward  and  forward  strokes  and 
remains  so  for  all  degrees  of  cut-off  up  to  mid-gear. 
The  gear  is  illustrated  in  Fig.  32.  The  use  of  eccen- 
trics is  dispensed  with.  At  a  point  A  in  the  con- 
necting rod  is  pivoted  a  link  B,  connected  at  its 
lower  end  to  the  radius  rod  C  which  restricts  its 
motion  to  a  vertical  plane.  The  point  A  on  the 
connecting  rod  describes  an  ellipse  when  working. 
At  the  point  D  a  lever  E  is  pivoted,  which  is  centred 
at  F  and  extended  to  the  point  K  where  con- 
nection is  made  with  the  lever  G.  The  path  of  the 
point  D  is  an  irregular  oval,  and  that  of  K  a  true 
vertical  ellipse.  The  valve  rod  V  is  pivoted  to  G. 
A  rising  and  falling  movement  is  communicated  to 
F  by  the  motion  of  the  connecting  rod,  its  movement 
being  guided  by  a  die  sliding  in  a  slot  J  which  has  a 
radius  of  curvature  equal  to  the  length  of  G.  G  rises 
and  falls  with  F  and  thus  communicates  a  horizontal 
movement  to  the  valve  spindle  V. 

The  block  in  which  the  slot  is  cut  is  capable  of 
pivoting  about  a  centre  F,  its  inclination  to  one  side 
or  the  other  being  effected  through  the  lever  L 
which  is  operated  from  the  reversing  screw  at  the 
foot-plate. 

The  degree  of  port  opening  and  consequently  the 
rate  of  expansion  is  regulated  by  the  inclination  of 
the  slot  from  the  vertical.  When  it  is  exactly  central 
as  shewn,  the  valve  is  in  mid-gear ;  when  thrown  over 


vi] 


UTILIZATION  OF  THE  STEAM 


109 


to  its  extreme  backward  or  forward  positions,  back- 
ward or  forward  running  of  the  engine  is  obtained. 

The  Walschaert  valve  gear,  which  is  preferred  on 
the  Continent,  has  also  a  constant  lead  for  all  points 
of  cut-oif  and  produces  a  more  uniform  steam  distri- 
bution than  the  Stephenson  gear. 


Fig.  33.     The  Walschaert  valve  gear. 

Referring  to  Fig.  33  it  will  be  seen  that  the 
movement  of  the  valve  is  derived  from  two  sources, 
the  crosshead  F  and  an  eccentric  crank  A,  whose 
centre  is  situated  90°  from  the  centre  line  on  the 
main  crank,  when  the  centre  lines  of  the  cylinders 
and  gear  motion  coincide  and  pass  through  the 
centre  of  the  axle.  From  the  eccentric  crank  A  an 
eccentric  rod  E  runs  and  makes  connection  with  the 
link  D  which  is  pivoted  in  the  centre. 


110  THE  MODERN  LOCOMOTIVE          [OH. 

A  groove  in  the  link  contains  a  die  P,  which  is 
free  to  slide  up  and  down  therein:  this  block  is 
attached  to  a  radius  rod  S,  the  length  of  which  be- 
tween the  points  D  and  L  is  equal  to  the  radius  of 
the  link  itself.  If  the  radius  bar  were  pivoted  directly 
to  the  valve  spindle  F,  when  the  crank  was  on  the 
centre,  the  valve  would  be  in  its  mid  position  for 
either  backward  or  forward  running  and  there  would 
be  no  lap  or  lead.  Lap  and  lead  are  obtained  by  a 
rigid  arm  6r,  dropped  from  the  crosshead  centre  F. 
Pivoted  to  6r  is  a  union  arm  H,  making  connection 
with  the  combination  lever  K.  This  is  pivoted  at 
the  point  L  of  the  radius  bar  and  prolonged  to  form 
a  connection  with  the  valve  spindle  at  M.  It  will  be 
seen  that  the  inclination  of  the  combination  lever  K 
will  be  the  same  at  the  end  of  the  stroke  regardless 
of  the  position  of  the  radius  rod  S ;  and  that,  there- 
fore, the  horizontal  displacement  of  the  point  M  and 
the  valve  spindle  will  be  the  same  on  either  side  of  a 
vertical  line  through  L.  This  horizontal  displace- 
ment is  equal  to  twice  the  sum  of  the  lap  and  the 
lead,  hence  the  latter  is  constant  for  all  points  of  cut- 
off. It  should  be  pointed  out  that  the  eccentric  and 
the  crosshead  tend  to  move  the  valve  in  opposite 
directions  during  the  first  half  of  each  stroke  and  in 
the  same  direction  during  the  last  half;  or,  in  other 
words,  they  work  in  opposite  directions  during  the 
first  and  third  quarters  of  a  revolution  of  the  crank 


vi]          UTILIZATION  OF  THE  STEAM         111 

starting  from  either  dead  point,  and  together  during 
the  second  and  fourth  quarters.  The  motion  derived 
from  the  crosshead  is  constant  and  is  not  subjected 
to  reversal  in  the  reversing  of  the  motion  of  the 
engine,  which  is  done  exclusively  by  a  change  in  the 
motion  imparted  by  the  eccentric  crank,  which  also 
controls  the  variation  of  the  points  of  cut-off. 

In  order  to  accomplish  this  the  motion  of  the 
eccentric  crank  is  transmitted  through  an  oscillating 
link  pivoted  at  its  centre  and  so  slotted  that  a  die 
attached  to  the  back  end  of  the  radius  bar  can  be 
moved  through  its  whole  length,  and  by  placing  this 
above  or  below  the  centre  a  reversal  of  the  engine 
will  be  obtained.  This  motion,  either  direct  or  in- 
direct, is  taken  up  by  the  radius  bar  and  carried  out 
to  the  combination  lever,  where  it  is  combined  with 
that  obtained  from  the  crosshead  and  the  resultant 
imparted  to  the  valve.  The  features  which  have 
brought  the  Walschaert  gear  into  such  extensive  use 
abroad  are  its  simplicity,  lighter  parts  and  accessi- 
bility, all  of  which  are  obtained  without  loss  of 
efficiency. 


112  THE   MODERN  LOCOMOTIVE         [CH. 

CHAPTER   VII 

FRAMES   AND    RUNNING  GEAR 

THE  engine  as  a  carriage  does  not  shew  so  much 
divergence  from  the  standard  type  of,  say,  twenty 
years  ago  as  the  other  elements,  consequently  the 
briefest  review  will  suffice. 

Frames.  The  boiler  is  carried  on  frames  forming 
a  chassis,  which  is  in  turn  carried  by  the  wheels 
(Fig.  5).  The  frames  consist  essentially  of  two  longi- 
tudinal members  connected  at  the  front  and  back  by 
buffer  plates.  They  are  also  stayed  transversely  by 
the  cylinder  casting,  motion  plate,  and  intermediate 
cross-stays. 

Frames  have  to  be  made  strong  enough  to  counter- 
act the  alternate  tensional  and  compressional  stresses 
set  up  by  the  steam  acting  on  each  end  of  the  cylinders 
alternately,  and  also  they  have  the  pull  of  the  engine 
to  transmit  to  the  draw-bar.  Locomotive  frames  are 
arranged  vertically  between  the  wheels  and  are  in- 
variably made  of  mild  steel  plates  about  1  in.  to  1 J  in. 
thick.  Owing  to  the  gauge,  the  distance  between 
them  is  limited  to  about  4  ft.  2  in. 

They  are  merely  strong  plates  shaped  out  to  take 
the  axles  of  the  wheels,  and  suitably  drilled  for  the 
attachment  of  all  the  engine  details.  They  are  rolled 


vn]       FRAMES  AND  RUNNING   GEAR        113 

from  Siemens-Martin  mild  steel  ingots.  An  average 
frame-plate  ingot  weighs  about  five  tons,  and  will 
make  two  plates.  Frames,  previous  to  about  1868, 
were  made  in  three  lengths  welded  together,  as  at 
that  time  machinery  did  not  exist  for  rolling  them 
in  one  length.  In  America,  Bavaria,  Austria,  and 
other  countries  plate  frames  have  in  recent  years 
been  displaced  by  bar  frames.  As  their  name  im- 
plies, these  are  usually  built  up  of  bar  sections  welded 
or  strongly  bolted  together  and  consist  generally  of 
two  main  portions,  a  front  section  supporting  the 
cylinders  and  motion  parts,  and  a  back  portion  which 
accommodates  the  axle-box  guides  or  pedestals.  Con- 
nection between  the  two  sections  is  made  by  two  arms 
forming  an  extension  of  the  front  pedestal,  between 
which  is  spliced,  bolted  and  keyed  the  front  portion 
of  the  frame.  The  length  of  frame  between  the 
leading  and  trailing  wheels  is  usually  doubled, 
the  upper  bar  or  'top  rail'  and  the  lower  member 
or  ' bottom  rail'  being  stayed  together  by  ribbed 
uprights,  which  are  utilized  for  suspending  the  brake 
hangers.  Sometimes  the  extension  passing  beneath 
the  cylinders  is  also  doubled,  but  for  large  engines 
the  two  bars  give  place  to  a  single  slab  to  which  the 
cylinders  are  bolted.  The  pedestals  are  stayed  across 
the  openings  by  pedestal-binders  equivalent  to  horn- 
stays.  The  cylinder  castings  are  relied  upon  to  bear 
the  greater  portion  of  the  burden  of  keeping  the 

A.  L.  8 


114  THE  MODERN  LOCOMOTIVE         [OH. 

frames  in  alignment,  the  remainder  of  the  staying 
being  obtained  by  broad,  well-ribbed  cross-ties  run- 
ning horizontally  and  diagonally  between  the  frames 
and  placed  as  close  to  the  pedestals  as  possible.  The 
frames  are  usually  of  wrought  iron  about  4  inches  in 
section.  It  is  of  interest  to  note  that  the  earliest 
locomotives  were  fitted  with  bar  frames :  subsequent 
engines  had  wood  frames  plated  with  iron  and  a  wood 
buffer  beam  at  each  end. 

Frame  failures  are  of  fairly  common  occurrence  in 
America,  and,  being  expensive  to  repair,  the  practice 
of  using  cast  steel  frames  having  an  I ,  instead  of  a 
rectangular  section,  is  being  extensively  introduced. 
From  theoretical  considerations  a  frame  of  this  type, 
allowing  the  same  amount  of  metal  and  the  same 
width  of  frame,  has  been  shewn  to  be  about  four 
times  as  strong  in  the  horizontal  plane  and  a  little 
over  half  as  strong  in  the  vertical  plane.  The  dis- 
advantage is  that  welding  is  difficult  if  the  frame  is 
broken. 

Advantage  may  be  taken  of  this  opportunity  to 
state  that  generally,  in  the  construction  of  loco- 
motives, steel  castings  are  replacing,  more  and  more, 
forged  pieces.  The  Belgian  State  Railway  not  long 
since  gave  builders  a  long  list  of  parts,  at  present 
forged,  which  they  are  allowed  to  replace  at  will  by 
steel  castings.  In  Hungary  foundation  rings  are  steel 
castings,  and  in  Germany  bar  frames  have  also  been 


vn]       FRAMES  AND  RUNNING  GEAR        115 

cast  of  steel.  Other  details  such  as  guide-bars,  cross- 
heads,  frame-braces,  stretchers,  smoke-box  saddle, 
buffers,  and  buffer-brackets,  together  with  brake- 
beams  and  a  number  of  lesser  parts,  have  also  been 
made  of  steel  castings.  It  is,  of  course,  not  necessarily 
a  cheap  engine  that  is  built  of  steel  castings,  but  it 
may  be,  and  generally  is,  much  lighter  than  if  forgings 
had  been  used.  To  resist  the  strains  of  traction  and 
of  buffing,  box-girder  frames  have  also  been  used  on 
heavy  goods  engines  on  the  Continent.  To  eliminate 
the  element  of  weakness  inherent  in  welded  or  bolted 
parts,  bar  frames  cut  from  the  solid  are  used  in  some 
cases.  The  writer  recently  saw  a  set  in  the  process 
of  manufacture  at  the  works  of  the  North  British 
Locomotive  Company.  They  were  being  made  from 
solid  steel  slabs  weighing  8  tons  each.  The  holes 
were  first  drilled,  then  the  slab  was  ripped  up,  planed 
and  finished  on  a  shaping  machine.  The  weight  of 
frame  finished  was  only  1  ton  19  cwts. ! 

A  combination  of  plate  and  bar  methods  has  also 
been  tried  on  foreign  locomotives,  the  rear  portion 
being  constructed  on  the  plate  system. 

Wheels.  The  wheel  comprises  tyre,  wheel-centre, 
axle  and,  in  a  coupled  wheel,  crank-pin.  The  tyres 
are  bored  out  somewhat  smaller  than  the  wheel 
centres,  and  are  shrunk  on  to  them,  a  usual  shrink- 
age allowance  being  j^  of  the  diameter  of  the  wheel 
centre.  The  tread  of  the  tyre  is  turned  up  when 

8—2 


116  THE  MODERN  LOCOMOTIVE         [CH. 

the  wheel  and  axle  are  finished,  and  is  left  rough- 
turned  to  assist  adhesion.  To  enable  the  wheel 
centre  to  be  placed  in  the  tyre,  the  latter  is  expanded 
in  a  gas  furnace  and  the  wheel  centre  lowered  into  it. 
Wheel  centres  were  at  one  time  forged,  but  now  are 
usually  cast  in  steel.  Some  railways  use  cast  iron  for 
the  wheel  centres  of  mineral  engines. 

Wheel  centres,  after  being  turned  and  bored,  are 
pressed  on  to  the  wheel  seat  of  the  axle  by  hydraulic 
pressure,  at  from  8  to  12  tons  per  inch  diameter  of 
axle. 

Several  methods  of  securing  the  tyre  on  the  rim 
are  in  vogue.  On  the  London  and  North  Western,  the 
outer  edge  of  the  tyre  is  turned  outwards  so  as  to 
form  a  recess  and  lip,  a  corresponding  projection 
being  formed  on  the  rim,  which  fits  into  the  recess. 
Set  screws  are  screwed  into  the  rim  at  intervals  with 
their  ends  projecting  into  the  rim. 

Axles.  A  straight  axle  consists  of  two  journals, 
two  wheel  seats  and  the  shank:  in  a  crank  axle 
the  two  cranks  occupy  the  place  of  the  shank.  The 
journals  are  case-hardened.  A  small  percentage  of 
chromium  is  nowadays  introduced  into  axle  steel,  as 
this  has  been  found  to  toughen  it.  Crank  axles  are 
either  forged  from  slabs  under  the  hammer-press  or 
built  up.  A  built-up  axle  consists  of  one  piece  of 
axle  for  each  end,  a  middle  piece,  four  crank  cheeks 
and  two  pins ;  they  are  machined  and  key  ways  cut 


vii]       FRAMES  AND  RUNNING  GEAR        117 

previous  to  assembling.  It  is  usual  to  shrink  the 
parts  together,  but  the  crank  cheeks  are  all  keyed 
to  their  respective  parts  of  the  axle  in  addition.  The 
oblique  arm-crank  axle  has  an  extended  application 
on  Belgian  and  some  French  locomotives.  This  is  a 
very  strong,  cranked  axle  and  is  often  made  hollow 
throughout  except  in  the  oblique  portion. 

Other  running  part  details.  Connecting  rods  are 
mild  steel  forgings,  planed,  bored  and  slotted  on  all 
working  faces,  i.e.  where  brasses  or  cottars  fit,  but 
milled  up  and  polished  elsewhere,  as  flaws  shew  up 
better  on  polished  surfaces.  The  small  ends  are 
usually  brass  lined  or  bushed  for  the  gudgeon  pin, 
and  the  big  ends  fitted  with  some  arrangement  of 
adjustable  brass. 

Coupling  rods  are  bushed  at  each  end  for  the 
crank-pin,  and  machined  out  to  a  channel  section,  as 
this  style  combines  lightness  with  strength.  Eccentric 
sheaves  and  straps  are  usually  of  cast  iron,  and  the 
rods  of  Yorkshire  iron.  The  straps  are  often  fitted 
with  a  removable  cast  iron  liner  which  can  be  easily 
renewed.  The  sheaves  are  keyed  and  also  held  by 
pinching  screws.  All  the  motion  pins,  etc.,  are  made 
from  mild  steel  and  are  case-hardened  and  ground  up 
true  after  machining.  Axle  boxes  are  steel  castings 
with  a  semi-circular  brass  strip  with  white  metal 
insets  called  a  '  step '  let  into  the  bearing  part  of  the 
casting  for  the  axle  journal  to  wear  against.  The 


118  THE  MODERN  LOCOMOTIVE         [CH. 

edges  of  the  box  are  planed  to  wear  against  the 
horn  block  faces.  These  horn  blocks,  or  axle  box 
guides  (cf.  Fig.  5),  are  of  cast  steel  and  are  fixed 
to  the  frames  by  rivets.  A  distance  piece  and  a 
strong  bolt  through  the  ends  of  the  horns  are  fitted 
to  stay  the  bottoms  of  the  horns  of  the  frame  to- 
gether. 

Radial  Axle  Boxes,  Bogies,  and  Bissels.  In  the 
earlier  stages  of  locomotive  building  it  was  customary 
to  rely  on  the  flange  of  the  driving  wheel  for  guidance 
and  to  force  the  engine  to  turn  when  entering  a  curve. 
With  low  speeds  this  was  sufficient,  especially  as 
English  engineers  kept  the  track  in  as  nearly  perfect 
condition  as  possible  and  made  these  curves  of  very 
large  radius.  With  the  increasing  length  of  engines 
some  guiding  device  became  imperative,  and  the 
radial  axle  box,  bogie  and  pony  truck  or  Bissel  were 
developed. 

No  better  example  of  a  radial  axle  box  exists 
than  that  designed  by  the  late  Mr  Webb  for  his 
engines  on  the  London  and  North  Western  Railway. 
In  this  a  certain  amount  of  side-play  is  secured  by 
uniting  the  two  axle  boxes  in  one  curved  casting 
which  is  capable  of  sliding  If  ins.  to  the  right  or  left 
from  its  central  position,  over  a  guide  curved  to  a 
corresponding  arc  of  circle.  The  wheel  and  casting 
are  brought  back  to  their  central  position  by  hori- 
zontal springs  as  the  engine  leaves  the  curve  and 


vii]       FRAMES  AND  RUNNING  GEAR        119 

runs  on  to  the  straight.  They  are,  however,  being 
displaced  from  their  position  of  leading  wheels  by  the 
superior  devices  known  as  the  bogie  and  pony  truck, 
although  they  are  coming  into  favour  with  Atlantic 
and  Pacific  type  engines,  as  trailing  wheels,  in  which 
the  trailing  axle  being  placed  from  9  to  12  ft.  behind 


Fig.  34.     Pony  truck  or  '  Bissel.* 

the  rear  coupled  axle,  must  necessarily  be  allowed 
the  same  'play.'  The  pony  or  Bissel  truck,  Fig.  34, 
is  used  when  the  weight  is  not  too  great  for  one  pair 
of  wheels.  It  is  pivoted  by  means  of  a  radius  bar  to 
some  point  on  the  frame  in  rear  of  its  axle,  the  whole 
truck  frame  being  free  to  slew  sideways  under  the 


120  THE  MODERN  LOCOMOTIVE         [CH. 

point  of  the  engine  except  as  controlled  by  swinging 
links  L,  these  being  intended  to  allow  the  turning 
forces  to  act  gradually  on  the  front  part  of  the 
engine.  Sometimes  horizontal  springs  are  used 
instead  of  swing  links.  Their  action  in  this  con- 
nection has  been  explained  above  in  dealing  with 
the  radial  axle.  The  swinging  links  are  attached 
at  their  top  ends  to  the  truck  frame  and  at  the 
lower  ends  to  brackets  on  the  socket  of  the  engine 
centre  pin.  On  entering  a  curve,  the  side  movement 
of  the  truck  causes  one  of  the  links  to  shorten  in  its 
effective  vertical  length.  This  has  the  effect  of  lifting 
the  engine  or,  what  is  the  same  thing,  putting  more 
pressure  on  the  springs  at  that  side,  which  pressure 
tends  to  bring  the  truck  back  to  its  central  position 
on  leaving  the  curve.  The  truck  is  so  designed  that 
the  engine  forces  the  centre  line  of  the  truck  to  take 
a  direction  parallel  to  the  tangent  at  the  point  on  the 
curve  where  the  truck  wheels  are  bearing.  This  is 
accomplished  by  pivoting  the  truck  at  such  a  distance 
from  the  truck  centre  pin  that  the  angle  0  through 
which  the  truck  turns,  will  be  greater  than  the  angle 
between  the  tangent  to  the  curve  where  the  truck 
bears  and  the  centre  line  of  the  engine.  The  outer 
truck  wheel  flange  always  bears  against  the  outer  rail 
when  6  is  less  than  this  angle  and  against  the  inner 
rail  when  6  is  greater.  The  exception  to  this  rule 
occurs  when  the  weight  on  the  truck  is  not  sufficient 


vn]       FRAMES  AND  RUNNING  GEAR        121 

to  prevent  the  truck  from  being  slid  against  the 
outside  rail. 

The  now  familiar  four-wheeled  truck  or  bogie,  for 
carrying  the  leading  end  of  the  engine,  was  introduced 
in  America,  and  for  a  long  time  was  regarded  with 
disfavour  by  English  engineers.  The  necessity  of 
securing  a  flexible  wheel  base  with  an  increasing 
weight  of  the  leading  end  of  the  engine  has,  however, 
ultimately  led  to  its  extensive  adoption.  As  first 
introduced  it  was  allowed  only  a  simple  rotating 
movement  about  its  centre  pin  upon  which  the  front 
end  of  the  locomotive  rested.  They  were  then  made 
with  swinging  links,  as  described  above,  so  as  to 
allow  a  certain  amount  of  swing  or  horizontal  play 
perpendicular  to  the  centre  line  of  the  truck.  In 
some  designs  the  links  are  replaced  by  spiral  or 
laminated  springs  on  each  side  to  secure  control, 
but  there  is  some  divergence  of  opinion  as  to  which 
method  is  preferable.  While  swinging  links  are  held 
to  possess  the  advantage  in  smoothness  of  action  and 
freedom  from  friction  they  are  stated  by  some  not  to 
give  enough  pressure,  and  this  can  only  be  secured 
by  the  use  of  springs. 

The  theory  of  the  action  of  the  bogie  and  the 
calculation  of  the  flange  pressure  and  stresses  set 
up  involve  a  somewhat  complicated  mathematical 
treatment  outside  the  scope  of  this  manual. 

Wheel  Arrangement.    The  wheel  arrangement  is 


122  THE   MODERN  LOCOMOTIVE         [CH. 

important  inasmuch  as,  in  conjunction  with  the  eva- 
porative capacity  of  the  boiler,  it  largely  determines 
the  type  and  function  of  the  locomotive.  According 
to  the  old  system  of  classification,  engines  were  di- 
vided into  four  classes,  namely,  (1)  express,  (2)  mixed, 
(3)  goods,  (4)  local  or  tank.  Further,  when  three 
pairs  of  wheels  were  coupled,  they  were  known  as 
' six-coupled '  engines;  four-coupled'  denoted  that 
two  pairs  of  wheels  were  connected  by  coupling  rods, 
and  when  one  pair  alone  was  used  for  driving,  the 
locomotive  was  designated  a  'single'  engine.  For 
many  years,  express  passenger  single  engines  enjoyed 
great  popularity  because  they  were  very  free  running 
and  capable  of  attaining  high  speeds  with  a  load 
suited  to  their  power. 

The  old  and  somewhat  confused  system  of  classifi- 
cation has  now  been  displaced  by  a  notation  which  is 
at  once  capable  of  indicating  explicitly  any  particular 
type  of  engine.  In  using  it  one  is  supposed  to  be 
standing  on  the  footplate  of  the  engine  and  looking 
ahead,  and  considering  in  succession  the  bogie  or 
leading  wheels,  driving  and  coupled  wheels,  and 
trailing  wheels.  Thus  a  'single'  engine  would  be 
designated  a  2-2-2  type;  another,  with  a  leading 
bogie  and  four  coupled  wheels,  would  be  represented 
in  the  notation  as  belonging  to  the  4-4-0  class  and 
so  on,  the  0  indicating  the  absence  of  any  trailing 
wheels.  A  four-wheel  coupled-in-front  engine  with 


vn]       FRAMES  AND  RUNNING  GEAR        123 


124  THE  MODERN  LOCOMOTIVE          [OH. 

no  leading  wheels  and  a  trailing  bogie  would  be 
designated  a  0-4-4  type.  The  principal  types  in 
favour  to-day  are  set  out  in  a  diagrammatic  chart, 
Fig.  35,  which  will  enable  the  method  to  be  readily 
followed. 

The  prevailing  types  of  the  modern  express  pas- 
senger engine  in  this  country  are  the  4-4-2  Atlantic 
and  the  4-6-0  types  (see  Figs.  18  and  20),  although  a 
large  number  of  the  older  2-4-0  and  4-4-0  types  are 
still  in  service.  The  Mogul  2-6-0  type  is  represented 
on  the  Great  Western  Railway.  The  0-6-0  and  0-8-0 
are  popular  types  of  heavy  goods  engines,  and 
examples  of  the  4-8-0  and  0-8-4  engines  are 
represented  by  heavy  shunting  engines. 

On  the  whole  it  may  be  said  that  the  Atlantic 
type  is  generally  the  most  popular  to-day,  chiefly  for 
the  reasons  that  it  permits  the  use  of  a  long  boiler 
barrel  with  an  augmented  volume  of  water,  large 
fire-box  and  a  high  evaporative  efficiency ;  com- 
paratively short  coupling  rods ;  an  advantageous 
distribution  of  weight ;  finally  a  small  fixed- wheel  base 
which  renders  the  engine  easy  on  curves.  Although 
it  may  mean  a  diminished  stability  if  the  trailing 
wheels  are  allowed  an  excessive  uncontrolled  radial 
movement,  it  stands  for  the  highest  degree  of  develop- 
ment of  the  four-coupled  engine  and  is  par  excellence 
the  type  of  modern  specialized  express  engine  for 
all  but  the  heaviest  loads  and  severe  gradients. 


vin]  STABILITY  125 

CHAPTER  VIII 

STABILITY 

THE  exterior  forces  acting  on  the  locomotive  as 
a  whole,  assuming  that  the  track  is  perfectly  straight 
and  horizontal,  are,  besides  the  weight,  the  effect 
due  to  the  resistance  of  the  train  and  the  air,  the 
tangential  reaction  of  the  rails  on  the  coupled  wheels, 
the  forces  due  to  the  inertia  of  the  reciprocating 
parts,  and  the  horizontal  component  of  the  centri- 
fugal force  of  the  revolving  masses  where  they  have 
not  been  completely  balanced.  Other  factors  tending 
to  set  up  perturbations  in  the  movement  of  the  engine 
are  variations  of  the  effort  on  the  piston,  and  the 
elasticity  of  the  draw  gear. 

The  disturbances  due  to  the  centrifugal  forces  and 
centrifugal  couples  set  up  by  the  revolving  masses  are 
such  as  to  develop  oscillations  in  the  engine  unless 
they  are  balanced  so  as  to  reduce  them  to  the  smallest 
possible  amount.  A  familiar  example  of  centrifugal 
force  occurs  when  a  stone  or  small  bullet  is  whirled 
round  at  the  end  of  a  long,  fine  string.  This  string 
itself  is  pulled  away  from  the  centre  by  the  bullet, 
which  is  said  to  exert  on  it  a  centrifugal  force.  A 
simple  extension  of  this  example  is  the  face  plate 
of  a  lathe  to  which  is  eccentrically  attached  a  heavy 


126  THE  MODERN  LOCOMOTIVE         [CH. 

piece  of  work.  If  this  is  revolved,  a  wobbling  or 
vibrating  motion  results  which  sets  up  stresses  in 
the  frame  and  bearings  varying  in  magnitude  as 
the  square  of  the  angular  velocity,  and  directly  as  the 
distance  between  the  axis  or  spindle  and  the  revolving 
mass.  If  continued  the  bearings  would  wear  rapidly 
and  unequally,  hence  when  work  of  this  kind  has  to 
be  dealt  with,  it  is  balanced  by  attaching  a  piece 
of  iron  of  equal  weight  to  the  opposite  side  of  the 
face  plate.  It  is  not  difficult  to  see  from  this  that 
a  crank  is  also  an  example  of  an  unbalanced  force. 

To  illustrate  a  centrifugal  couple,  let  a  cord  be 
attached  to  the  centre  of  a  stick  and  whirled  round. 
It  will  be  found  to  always  set  itself  radially  to  the 
axis  of  revolution,  which  is  due  to  the  centrifugal 
couple  set  up. 

A  couple,  it  may  be  explained,  is  the  name  given 
to  a  pair  of  equal  and  opposite  forces  acting  in 
parallel  lines. 

Two  equal  masses,  such  as  two  cranks  set  at  180° 
to  each  other  and  revolving  in  different  planes  on  the 
same  shaft,  furnish  another  example.  Each  develops 
a  centrifugal  force  equal  in  magnitude  and  opposite 
in  sign,  and  form  a  couple  tending  to  turn  the  shaft  in 
the  plane  of  the  axis  of  revolution.  But  it  is  not  only 
the  centrifugal  forces  set  up  by  the  rotating  cranks 
that  require  to  be  balanced,  there  are  also  to  be 
considered  the  disturbing  effects  of  the  reciprocating 


vm]  STABILITY  127 

parts  such  as  the  piston,  piston  rod  and  crosshead ; 
also  those  due  to  the  motion,  varying  from  a  straight 
line  to  a  circle,  of  the  connecting  rod. 

Briefly,  the  principle  by  which  such  a  system  is 
established  more  or  less  in  equilibrium  is,  to  quote 
Rankine,  '  to  conceive  the  mass  of  the  piston,  piston 
rods  and  connecting  rods,  and  a  weight  having  the 
same  statical  moment  as  the  crank,  as  concentrated 
at  the  crank  pins  and  to  insert  between  the  spokes  of 
the  driving  wheels  counterpoises.'  The  weights  and 
position  of  these  balancing  masses  can  be  approxi- 
mately determined  by  calculation.  Balancing  the 
revolving  masses,  such  as  crank-arms,  crank-pins, 
etc.,  offers  a  problem  of  no  special  difficulty.  It  is 
accomplished  either  by  prolonging  the  crank-arms 
to  the  opposite  side  of  the  axle  to  form  a  balance 
weight,  or  by  putting  balance-weights  on  the  inside 
of  the  wheel  rims.  The  first  method  avoids  the 
centrifugal  couple  between  the  wheels,  which  sets  up 
a  bending  moment  on  the  axle.  The  last-named 
method  is  the  most  favoured.  It  was  used  as  far 
back  as  1842,  when  MacConnell,  at  the  suggestion  of 
Heaton,  a  Birmingham  engineer,  adopted  balance 
weights  on  the  Birmingham  and  Gloucester  Railway. 

The  problem  of  balancing  the  reciprocating  parts 
consisting  of  the  piston,  piston  rod  and  crosshead, 
and  that  portion  of  the  connecting  rod  which  reci- 
procates, is  not  so  simple.  For  a  full  discussion  of 


128  THE  MODERN  LOCOMOTIVE         [CH. 

the  question  the  reader  is  referred  to  Prof.  Dalby's 
standard  treatise  on  the  subject.  In  this  place  we 
can  only  summarize  some  of  his  results. 

Where 
M  is  the  mass  in  pounds  of  the  reciprocating  parts 

belonging  to  each  cylinder ; 
r  the  crank  radius  in  feet ; 
n  the  revolutions  of  the  crank  axle  per  second ; 
d  the  distance  in  feet  between  the  cylinder  centre 

lines ; 
t  the  distance  between  the  centre  line  of  the  driving 

axle  and  the  line  of  traction. 
The  unbalanced  reciprocating  parts  cause 

(1)  An  unbalanced  force,  the  maximum  value 
of  which  is  given  by  17  Mn2r  Ibs.  weight.     This  force 
accelerates  the  whole  mass  of  the  train  positively  and 
negatively  in  the  direction  of  travelling.     The  con- 
siderable efforts  set  up  by  this  couple  react  upon 
the  draw-gear  of  the  locomotive,  and  give  rise  to  a 
jerky  motion  which  severely  tests  the  strength  of  the 
engine  frames. 

(2)  A  couple  whose  maximum  value  is 

0'85  Mn*rd  foot-lbs. 

This  couple  produces  an  oscillatory  motion  about  a 
vertical  axis,  which,  superposed  upon  the  general 
forward  motion  of  the  engine,  causes  a  swaying  from 
side  to  side.  This,  when  acting  on  a  short  engine, 
may  become  dangerous  at  high  speeds. 


vin]  STABILITY  129 

(3)    A  couple  whose  maximum  value  is 

17  Mn2rt  foot-lbs. 

This  couple  tends  to  cause  oscillation  in  a  vertical 
plane  about  a  horizontal  axis.  The  reader  can  work 
out  the  values  for  himself  from  the  following  data 
for  weights:  connecting  rods,  reciprocating  portion, 
180  Ibs.;  piston,  150  Ibs.;  piston  rod,  crosshead,  and 
pin  cottar  and  nut,  180  Ibs. 

The  best  method  of  balancing  reciprocating  parts 
is  not  immediately  obvious  ;  fortunately,  however,  it 
can  be  shewn  that  although  they  move  in  straight 
lines  or  describe  ovals  in  a  vertical  plane,  it  comes 
to  the  same  thing  if  they  are  considered  as  a  body 
of  equal  mass  revolving  in  a  circle  whose  centre  is 
the  crank  axle.  Therefore  in  a  two-cylinder  engine 
the  reciprocating  force  and  couple  are  dealt  with 
in  the  same  way  as  those  due  to  the  revolving  masses, 
namely,  by  placing  balance  weights  inside  the  rim  of 
the  driving  wheel.  Separate  weights  are  not  used 
for  each  set  of  parts,  but  are  combined. 

Unfortunately  in  curing  one  trouble  another  is 
created,  for  the  revolving  balance-weights  themselves 
set  up  a  centrifugal  force,  which  acts  at  right  angles 
to  the  plane  in  which  the  reciprocating  parts  move. 
The  horizontal  component  of  the  force  tends  to  thrust 
the  axle  box  against  the  guides,  and  the  vertical  com- 
ponent acts  by  lifting  the  wheel  at  one  instant  and 
dropping  it,  thus  causing  on  the  one  hand  slipping, 

A.  L.  9 


130  THE  MODERN  LOCOMOTIVE         [CH. 

and  on  the  other  a  'hammer  blow'  action  on  the 
rail. 

This  hammer  blow  can  amount  to  as  much  as 
25  per  cent,  of  the  static  weight,  and  is  not  only 
injurious  to  rails,  tyres  and  bridges,  but  limits  the 
carrying  capacity  of  the  axle.  The  locomotive 
engineer  is  thus  on  the  horns  of  a  dilemma :  either 
he  can  fully  balance  the  reciprocating  parts  and  so 
eliminate  the  swaying  couple ;  or  he  can  leave 
them  unbalanced,  doing  away  with  hammer  blows 
and  leaving  the  engine  to  lurch  along  the  road  with 
the  certainty  of  derailment  if  very  high  speeds  are 
reached. 

Practice  has  regulated  the  amount  of  compromise, 
which  is  obviously  the  only  way  out,  by  balancing 
completely  the  revolving  parts  and  only  three-fourths 
of  the  weight  of  the  reciprocating  parts. 

The  connecting  rod,  which  partly  revolves  and 
partly  reciprocates,  cannot  be  perfectly  balanced  by 
a  rotating  weight,  but  can  be  balanced  in  a  vertical 
direction  by  dividing  the  masses  of  the  rod  between 
the  crank-pin  and  crosshead.  The  obliquity  of  the 
connecting  rod  is  also  responsible  for  a  variation  in 
rail  load  which,  compared  with  the  effect  of  the 
balance  weights,  may,  however,  become  negligible 
at  high  speeds. 

The  force  acting  at  the  crank  is  accompanied  by  a 
reaction  on  the  slide  bar,  and  forms  a  couple  tending 


vin]  STABILITY  131 

to  move  the  frame  in  a  direction  in  which  the  forces 
act  and,  as  the  locomotive  is  spring  borne,  to  vary 
the  load  on  the  springs.  Its  magnitude  depends 
upon  the  speed,  the  amount  cut-off  in  the  cylinder 
and  the  strength  of  the  springs,  and  the  relative 
length  of  the  rod  and  crank. 

The  problem  of  balancing  is  more  readily  solved 
with  four-cylinder  engines  in  which,  by  setting  the 
cranks  at  180°,  the  reciprocating  parts  are  auto- 
matically placed  in  equilibrium  and  need  no  balance 
weights.  This  disposition,  however,  introduces  costly 
complications  by  the  fact  that  each  cylinder  requires 
the  use  of  an  independent  valve  gear.  A  practical 
way  out  of  the  difficulty  is  secured  by  arranging  the 
cranks  in  pairs  at  180°,  the  pairs  themselves  being 
at  90°.  The  reciprocating  masses  are  then  balanced 
except  for  a  horizontal  swaying  couple  which  is  in- 
considerable. An  excellent  example  of  an  engine 
balanced  in  this  manner  is  furnished  by  Mr  Hughes' 
six- wheel  coupled,  four-cylinder  passenger  engine  on 
the  Lancashire  and  Yorkshire  Railway.  The  outside 
reciprocating  parts  weigh,  roughly,  595  Ibs.,  and  the 
inside  parts  524  Ibs.,  so  that  they  are  for  all  practical 
purposes  balanced  by  the  cranks  being  set  as  men- 
tioned above.  The  outside  revolving  masses  total 
1 108  Ibs.  and  those  inside  1255  Ibs.  These  are  balanced 
by  the  method  already  mentioned  of  prolonging  the 
crank-arms  to  form  balance  weights.  The  connecting 

9—2 


132  THE  MODERN  LOCOMOTIVE         [CH. 

rod  is  also  dealt  with  as  stated  earlier.  There 
remains  to  be  balanced,  the  unbalanced  portion  of 
the  crank  boss,  part  of  the  crank-pin  and  coupling 
rod,  which  are  dealt  with  by  balance  weights  placed 
in  the  leading  and  trailing  parts  of  driving  wheels. 

The  tangential  reaction  of  the  rails  on  the  driving 
wheels  involves  a  consideration  of  the  torsion  set 
up  in  the  driving  axle.  If  the  engine  has  outside 
cylinders,  the  effort  set  up  on  the  right  hand  crank 
pin,  for  example,  can  only  be  transmitted  to  the  left 
hand  wheel  by  a  torsional  stress  in  the  axle,  and 
whatever  its  magnitude,  this  implies  a  corresponding 
backward  slipping  of  the  right  hand  wheel,  which, 
however,  cannot  take  place  so  long  as  the  limit  of 
adhesion  has  not  been  passed.  In  the  case  of  an  inside 
cylinder  engine,  the  effort  on  each  piston  is  trans- 
mitted to  the  wheels  in  inverse  proportion  to  the 
distance  between  its  centre  line  and  the  wheel.  The 
effort  on  the  piston  also  sets  up  an  effort,  call  it  P, 
in  an  opposite  sense  on  the  cylinder-fastening  and 
therefore  on  the  main  frame  :  also  the  effort  set  up 
by  the  crank  sets  up  a  corresponding  reaction  in 
the  axle  boxes  p,  so  that  the  real  effort  exercised  by 
the  engine  is  the  difference  P  -p.  Alternating  stresses 
thus  occur  in  the  frames  between  the  cylinders  and 
the  crank-axle  which  may  be  sufficiently  high  to 
cause  weakening  of  the  cylinder  attachments  and 
sometimes  rupture  of  the  frames.  Another  effect  is 


vni]  STABILITY  133 

to  set  up  a  pressure,  first  in  one  direction  and  then 
in  another,  between  the  journals  and  bearings,  and 
between  the  axle  boxes  and  their  guides  which 
causes  a  '  knocking '  when  there  is  any  play  between 
the  parts. 

All  who  are  connected  with  railways  know 
that  certain  types  of  locomotives  when  driven  at 
high  speeds,  oscillate  vertically  and  laterally  to  a 
dangerous  extent  even  on  a  good  road.  Locomotives 
coupled  in  front  with  a  bogie  under  the  footplate 
(0-4-4  type)  are  held  by  some  to  be  specially  undesir- 
able for  high  speed  running.  An  example  of  this  was 
furnished  in  1895  by  the  Doublebois  accident  on  the 
Gt  Western  Railway,  in  which  two  such  locomotives 
hauling  a  fast  passenger  train,  left  a  practically  new 
road,  with  easy  curves.  Single  engines  with  a  single 
pair  of  leading  wheels  develop  a  considerable  sway- 
ing and  plunging  motion  which  the  introduction  of 
the  bogie,  however,  largely  removed. 

There  remains  to  examine  the  load  variation  on 
the  wheels  and  springs  set  up  during  running  by 
imperfections  of  the  road,  and  which  may  increase 
to  a  very  considerable  extent  any  tendency  to  sway- 
ing or  side  motion  of  the  engine  set  up  by  its  own 
action. 

An  oscillatory  or  galloping  movement  about  a 
transverse  axis  of  the  engine  is  produced  by  the 
rail  joints  or  inequalities  in  the  surface  of  the  rail. 


134  THE  MODERN  LOCOMOTIVE         [OH. 

Readers  will  doubtless  have  noticed  during  the 
passage  of  a  train  a  depression  at  the  joints  which 
results  in  the  rail  presenting  to  the  passage  of  a 
locomotive  a  curved  form,  the  lowest  point  of  which 
is  at  the  joint.  The  difference  of  level  in  the  rail 
itself,  according  to  M.  Coliard  may  reach  4  mm.  on 
good,  and  8  to  10  mm.  on  imperfect  roads.  The 
oscillations  so  set  up  do  not  reach  a  dangerous  pro- 
portion, except  when,  by  being  added  to  each  other, 
their  amplitude  goes  on  increasing.  This  increase 
takes  place  if  the  interior  friction  of  the  springs  does 
not  suffice  to  deaden  the  oscillations,  and  the  period 
becomes  that  due  to  the  disturbing  force,  namely,  the 
length  of  a  rail.  The  use  of  compensation  levers 
between  the  springs — a  standard  practice  in  France 
and  America — probably  has  a  considerable  influence 
on  this  type  of  oscillation. 

Oscillations  about  a  longitudinal  axis  situated  in 
the  plane  of  the  crank  axle  are  set  up  by  difference 
in  the  level  of  the  two  rails.  They  become  permanent 
when  the  joints  are  set  one  in  front  of  the  other,  but 
are  damped  out  by  the  interior  friction  of  the  springs. 
When  the  joints  are  directly  opposite  to  each  other, 
similar  oscillations  are  also  set  up  at  non-symmetrical 
points  in  the  rails,  at  crossovers,  and  under  certain 
circumstances  at  the  entrance  to  a  curve.  In  the  latter 
case  they  are  due  to  the  application  of  centrifugal 
force.  This  is  a  convenient  place  wherein  to  examine 


vin]  STABILITY  135 

the  conditions  necessary  for  an  engine  satisfactorily 
to  negotiate  a  curve. 

A  very  considerable  effort  is  required  to  keep  an 
engine  on  the  rails  and  prevent  it  persisting  in  a 
straight  course  according  to  Newton's  first  law  of 
motion.  This  eifort  is  given  by  the  formula  for 
centrifugal  force  acting  in  a  horizontal  direction  at 
the  centre  of  gravity  of  the  engine,  as  follows  :  — 


where 
p 
—  =  the  mass,  i.e.  weight  divided  by  the  force  of 

gravity, 

F=the  velocity  in  feet  per  second, 

R  =  the  radius  of  the  curve. 

In  early  locomotives  it  was  considered  important 
to  keep  the  centre  of  gravity  low,  but  more  recently 
it  has  been  seen  that  a  high  centre  of  gravity,  within 
certain  limits,  contributes  to  steadiness  of  running 
and  diminishes  the  lateral  pressure  upon  the  outer 
rail  in  curves,  by  reducing  the  obliquity  of  the  line 
of  thrust  on  the  rail  and  tending  to  make  it  more 
vertical.  The  load  on  the  outside  wheel  is  increased 
and,  therefore,  the  resistance  to  derailment.  The 
height  of  the  boiler  has  been  increased  from  5  ft.  3  ins. 
in  early  engines,  to  7  ft.  11  ins.  and  more  to  allow 
larger  boilers  to  be  used  with  large  diameter  driving 


136          THE  MODERN  LOCOMOTIVE         [OH. 

wheels.  It  is  important  to  notice  that  as  the  boiler 
forms  less  than  half  the  total  weight  of  the  locomotive, 
the  centre  of  gravity  of  the  whole  engine  is  only 
raised  by  less  than  half  the  amount  the  boiler  is 
raised.  Beyond  a  certain  limit,  however,  raising  the 
centre  of  gravity  increases  the  liability  of  the  engine 
to  overturn  by  increasing  the  overturning  moment 
due  to  centrifugal  force. 

Two  simple  calculations  based  on  actual  data 
derived  from  the  American  boat  train  disaster  at 
Salisbury  Station  in  1906,  will  suffice  to  illustrate 
the  approximate  condition  of  stability  of  a  modern 
locomotive  at  speeds  of  30  miles  and  70  miles  per 
hour  respectively.  The  following  are  the  necessary 
data : — 

( W)  Weight  of  locomotive    54  tons, 

( V)   Velocity  (30  miles  an  hour)    44  ft.  per  sec. 
(F)    Velocity  (70  miles  an  hour)  102'66  ft.  per  sec. 

(g)     Force  of  gravity     32*2  ft.  per  sec. 

(R)    Radius    of    the    curve    at 

Salisbury    523ft. 

(h)     Height  of  centre  of  gravity 

of  engine  above  rail  level  58 '5  ins. 
(I)  Horizontal  distance  between 
vertical  line  through  cen- 
tre of  gravity  of  engine 
and  outer  rail,  allowing 
for  3 Jin.  super-elevation  31 75 ins. 


vin]  STABILITY  137 

The  speeds  of  the  train  just  before  reaching 
Salisbury  were  worked  out  by  Mr  Holmes,  the 
superintendent  of  the  line  on  the  L.  and  S.  W.  R  as 
follows : — 

Length         Speed  in 

in  Miles 

Section  of  Line  Miles  per  hour 

Tisbury  to  Dinton      4-29  64-3 

Dinton  to  Wilton        5 '83  69 '9 

Wilton  to  Salisbury  (West)  ...        2-29  68 '5 

It  should  be  pointed  out  that  for  about  half  the 
distance  beyond  Wilton  there  is  an  up  gradient 
naturally  causing  diminution  of  speed,  while  the 
remainder  of  the  section  is  on  a  down  gradient  of 
1  in  115,  so  that  to  account  for  the  average  speed  of 
68*5  miles  an  hour  the  train  probably  travelled  at  the 
rate  of  at  least  70  miles  an  hour  between  the  west 
box  and  the  point  where  the  accident  occurred. 

The  moment  of  the  weight  of  the  locomotive 
about  the  outer  rail  is  obviously 

W  x  I  =  54  x  3175  =  1714'5  inch-tons, 

and  so  long  as  this  exceeds  the  overturning  moment 
due  to  centrifugal  force,  the  stability  of  the  engine 
will  be  assured,  but  of  course  it  may  be  in  a  critical 
condition  in  this  respect  unless  an  ample  margin  of 
safety  exists. 

At  the  velocity  of  30  miles  an  hour,  and  using 


138  THE  MODERN  LOCOMOTIVE         [CH. 

the  formula  given  above,  the  centrifugal  force  (/) 
developed  is 

,     W  x  r8       54  x  442 
/=  J^R  =  32-2x528  =  '*15  tOnS' 
and  the  moment  of  the  force  about  the  outer  rail  is 
/x  h  =  6'15  x  58'5  =  35977,  say  360  inch-tons. 

Comparing  this  value  with  that  of  the  moment  of 
stability  of  the  engine,  it  is  evident  that  at  the  speed 
of  30  miles  an  hour,  a  very  ample  factor  of  safety 
exists. 

At  the  velocity  of  70  miles  an  hour,  the  centri- 
fugal force  developed  is 

,     54  x  102-662 


32528 

and  the  moment  of  the  force  about  the  outer  rail  is 
/x  h  33'5  x  58'5  =  1960  inch-tons, 

which,  being  in  excess  of  the  moment  of  stability  of 
the  engine,  shews  that  at  the  speed  of  70  miles  an 
hour  round  the  curve  in  question,  it  is  impossible  for 
the  engine  to  avoid  being  overthrown. 

By  calculations  similar  to  the  above  it  would  be 
easy  to  demonstrate  that  the  velocity  giving  rise  to 
an  overturning  moment  exactly  equal  in  value  to 
that  of  the  moment  of  stability  for  the  engine  and 
curve  here  considered  is  equal  to  the  speed  of  about 
66  '2  miles  an  hour. 


vm]  STABILITY  139 

To  add  to  the  safety  of  engines  in  running  round 
curves,  super-elevation  of  the  outer  rail  is  resorted 
to,  the  amount  being  given  by  the  formula 

F2 

Gr25R> 
where 

G  =  Gauge  in  feet, 
F=  Velocity  in  miles  per  hour, 
R  =  Radius  of  the  curve  in  feet, 
E  =  Super-elevation  in  inches. 

On  the  Salisbury  curve  the  super-elevation  was 
3|  in.  and  therefore  suited  to  a  speed  of  only  34  miles 
per  hour,  and  according  to  the  working  time-sheets 
the  speed  over  the  curve  for  non-stopping  trains 
should  not  have  exceeded  25  miles  per  hour.  The 
effect  of  super-elevation  is  to  throw  the  centrifugal 
component,  acting  downwards,  well  within  the  gauge. 
If  it  falls  outside,  the  engine  is  unstable. 

There  is  yet  one  other  disturbing  effect  to  be 
noticed,  that  is  a  jerking  movement  from  side  to 
side  set  up  by  defects  in  the  road,  the  coning  of  the 
tyres  and  the  flexibility  of  the  suspension  system. 
Suppose  the  engine  to  be  deviated  to  the  left  a 
lateral  force  is  set  up,  causing  abrupt  contact  be- 
tween the  flange  and  rail,  and  resulting  in  a  more 
or  less  violent  shock.  If  the  rail  is  sufficiently  elastic 
the  force  will  be  completely  neutralized,  if  not  it 


140  THE  MODERN  LOCOMOTIVE         [CH. 

will  be  more  or  less  completely  given  back  to  the 
wheels,  which  will  result  in  an  impulse  towards  the 
opposite  rail.  Assisted  by  the  coned  shape  of  the 
tyres,  a  sinuous  movement  is  thus  set  up  which, 
unless  any  further  disturbance  is  set  up,  will  gradually 
lessen  until  the  action  of  the  rails  restores  normal 
running. 

Axle  play,  wear  of  the  tyres,  coupling  arrange- 
ments, a  swaying  couple  and  other  factors  modify 
the  circumstances  of  this  movement.  When  axles 
are  given  an  amount  of  play  in  relation  to  the  main 
frames,  they  may  take  a  sinuous  movement  inde- 
pendent to  that  of  the  engine  itself.  For  example, 
in  the  case  of  a  bogie  it  has  been  found  that  if  the 
lateral  control  is  insufficient,  it  will  float  along  or 
'get  across'  the  road.  The  bogie  as  well  as  the 
Bissel  however  possesses  the  great  advantage  of 
attenuating  shocks  and  so  relieves  the  stress  on  the 
permanent  way.  Although  their  mass  is  relatively 
small  and  their  lateral  reaction  equal  only  to  the 
tension  of  the  controlling  springs  or  pressure  on 
the  swinging  links,  they  are  capable  of  exerting  a 
powerful  leverage  tending  to  straighten  out  the 
running  of  the  engine. 

One  more  feature  of  the  bogie  may  be  mentioned, 
that  is,  its  tendency  to  flatten  the  road  for  the  driving 
wheels.  When  a  wheel  passes  over  a  particular  point 
in  a  rail,  that  point,  owing  to  the  elasticity  of  the 


ix]          PERFORMANCE  AND  SPEEDS          141 

rail,  is  depressed,  and  as  it  is  depressed  the  upward 
pressure  of  the  rail  is  increased.  As  the  pressure  on 
the  rail  is  removed,  the  upward  pressure  diminishes, 
but  not  so  quickly  as  it  had  increased.  In  other 
words  there  is  a  '  lag '  effect.  Thus  the  wheels  of  the 
bogie  carrying  the  weight  of  the  front  portion  of  the 
engine  would  set  up  a  deflection  in  the  rail  which, 
owing  to  the  lag,  the  driving  wheels  would  not  have 
to  repeat.  The  late  Mr  Webb,  in  a  discussion  at  the 
Institution  of  Civil  Engineers,  stated  that  he  well 
recollected  Mr  Patrick  Stirling  saying  one  day,  when 
he  had  been  discussing  with  him  how  he  had  managed 
to  pull  the  heavy  trains  on  the  Great  Northern  Rail- 
way with  single  driving  wheels,  'Mon,  I  have  the 
weight  on  the  bogie,  and  it  lays  the  road  down  for 
the  single  wheel  to  get  hold  of  it.' 


CHAPTER   IX 

PERFORMANCE  AND   SPEEDS 

THE  accurate  determination  of  the  factors  which 
enter  into  the  efficiency  of  the  locomotive,  although 
of  the  utmost  importance  in  practice,  for  a  long  time 
remained  a  matter  of  rule  of  thumb.  Rough  and 
ready  running  tests  no  doubt  contributed  a  great 
deal  of  valuable  information  to  the  locomotive 


142  THE  MODERN  LOCOMOTIVE          [OH. 

engineer,  but  it  was  not  until  quite  recent  years  that 
the  scientific  precision  adopted  in  stationary  engine 
testing  came  to  be  applied.  With  the  advent  of  the 
dynamometer  car  and  experimental  testing  plants 
means  have  been  placed  at  the  disposal  of  the  loco- 
motive engineer  for  determining  such  points  as  the 
average  rate  of  fuel  consumption,  steam  consumption 
at  various  speeds,  the  average  indicated  horse-power, 
the  effect  of  various  cut-offs  as  shewn  by  indicator 
diagrams,  smoke  gas  analysis,  resistance,  drawbar 
pull,  etc. 

Such  information  may  be  ascertained  in  two  ways, 
first,  under  actual  service  conditions  on  the  road, 
using  a  dynamometer  car;  second,  under  laboratory 
conditions,  with  the  engine  stationary  and  the  use  of 
a  testing  plant.  The  latter  method  gives  very  exact 
results  which,  however,  are  of  value  only  in  so  far  as 
they  are  tested  by  actual  running  experiments. 

For  conducting  the  latter  the  engine  is  fitted  up 
to  enable  indicator  diagrams  to  be  taken,  and  a 
dynamometer  car  is  included  in  the  train  between 
the  tender  and  first  vehicle.  The  use  of  the  latter 
will  be  understood  by  briefly  describing  the  equip- 
ment of  such  a  car  on  the  North  Eastern  Railway. 
The  body  of  the  car  is  built  on  a  steel  underframe 
shaped  to  take  a  special  spring  consisting  of  thirty 
steel  plates,  and  each  separated  by  rollers  to  minimise 
friction.  From  the  spring  the  pull  is  transmitted  to 


ix]          PERFORMANCE  AND  SPEEDS          143 

the  train.  As  the  spring  is  deflected  it  moves  a 
stylograph  pen  over  a  roll  of  paper,  thus  producing 
a  curve  of  drawbar  pull.  The  paper  is  caused  to 
travel  over  a  table  by  drums  driven  from  a  measuring 
wheel  which  rolls  on  the  rail,  thus  enabling  the 
operator  to  determine  the  speed  of  the  train  at  any 
instant.  The  permanent  speed  record  is  given  by  a 
pen  in  electrical  communication  with  a  clock,  which 
makes  a  mark  on  the  travelling  roll  of  paper  at  two- 
second  intervals.  The  speed  can  be  read  off  from 
this  by  the  aid  of  a  special  scale.  Dials  shew  the 
distance  travelled.  A  boiler  pressure  recorder  is 
also  fitted,  and  a  meter  is  constructed  on  the  same 
principle  as  a  planimeter  for  registering  the  work 
done.  In  this  apparatus,  a  horizontal  circular  plate 
moves  a  proportional  distance  to  that  of  the  train, 
whilst  a  frame  supporting  a  small  wheel  on  edge 
moves  across  it  from  the  centre  a  distance  propor- 
tional to  the  pull  on  the  drawbar.  Its  revolutions 
'are  therefore  a  measure  of  the  work  done,  and  as  it 
is  in  electrical  communication  with  a  meter,  the  work 
is  recorded.  An  indicator  records  the  pressure  in  the 
steam  chest,  and  another  registers  the  velocity  of  the 
wind,  which  blows  down  a  tube  kept  facing  its  direc- 
tion, and  causes  the  rise  and  fall  of  a  pen  on  the 
paper  drum.  The  direction  of  the  wind  is  indicated 
by  means  of  a  dial  in  the  roof. 

Most  British  railways  have  dynamometer  cars, 


144  THE  MODERN  LOCOMOTIVE         [OH. 

but  the  testing  plant  possessed  by  them  for  testing 
locomotives  under  laboratory  conditions  is  not  con- 
spicuous for  its  high  value  and  consists  chiefly  of 
friction  rollers.  In  1890  the  Purdue  University, 
Indiana,  put  down  the  first  plant  permitting  tests  to 
be  carried  out  on  really  scientific  lines  whose  example 
was  followed  by  the  Pennsylvania  Railroad  and  later 
by  the  Swindon  Works  of  the  Great  Western  Railway. 
With  these  plants  the  locomotive  to  be  tested  is 
placed  on  a  system  of  rollers  whose  centre  lines  are 
directly  underneath  those  of  the  locomotive  axles. 
It  is  kept  in  this  position  of  unstable  equilibrium  by 
a  very  ingenious  elastic  coupling,  which  measures 
the  tractive  effort  of  the  locomotive  at  its  wheels, 
while,  at  the  same  time,  it  prevents  any  displacement 
which  could  endanger  its  equilibrium.  The  rolling 
resistance  which  results,  when  running,  from  the 
resistance  proper  of  the  locomotive  and  that  of  the 
train  which  is  being  hauled,  is  produced  by  a 
hydraulic  brake  acting  on  the  supporting  rollers :  in 
consequence  of  this  action,  these  oppose  to  the 
rotation  of  the  driving  axles  a  resistance  similar  to 
the  reaction  of  the  rail.  A  revolution  counter  shews 
at  every  moment  the  speed  the  locomotive  would 
have  if  running  on  an  ordinary  track.  It  is  easy  to 
see,  how,  under  these  conditions  it  is  possible  to 
make  certain  experiments  which  it  would  be  very 
difficult  to  carry  out  while  running  on  the  track,  such 


ix]          PERFORMANCE  AND  SPEEDS          145 

as,  for  instance,  the  measurement  of  the  amount  of 
coal  which  can  be  burnt  per  square  foot  of  grate  and 
per  hour,  the  consumption  of  steam  at  different 
speeds,  and  for  different  settings  of  the  valve  gear. 
As  previously  stated  it  has  the  disadvantage  of 
placing  the  locomotive  under  artificial  conditions, 
and  the  most  serious  defect  is  the  impossibility  of 
studying  two  important  elements,  namely  the  ad- 
hesion and  the  rolling  resistance  of  the  locomotive. 
As  examples  of  some  of  the  results  obtained  with  the 
Pennsylvania  and  Purdue  plants  which  may  probably 
be  safely  applied,  the  following  out  of  a  large  number 
will  be  useful  for  reference : 

1.  When  working  at  maximum  power  the  boilers 
tested  generated  12  pounds  of  steam  per  square  foot 
of  heating  surface  per  hour. 

2.  The  evaporative  efficiency  falls  as  the  rate  of 
evaporation  increases.     When  working  at  full  power 
between  6  and  8  pounds  of  water  per  pound  of  dry 
'coal  were  evaporated. 

3.  Fire-box  temperatures  according  to  the  rate 
of  combustion  range  from  1400°  F.  up  to  2300°  F.  and 
smoke-box  temperatures  from  500°  F.  up  to  700°  F. 

4.  One    indicated    horse-power    per    hour   was 
developed  with  a  steam  consumption  of  from  23*8  to 
29  Ibs.  in  a  simple,  and  from  18*6  to  27  in  a  compound 
engine.     It  varies  with  the  speed  and  cut-off. 

5.  A  steam  locomotive  can  deliver  one  horse- 

A.  L.  10 


146  THE   MODERN  LOCOMOTIVE          [CH. 

power  at  the  drawbar  on  a  consumption  roughly  of 
about  2  Ibs.  of  coal. 

6.  The  mean  effective  pressure  varies  with  the 
speed  and  cut-off,  e.g.  speed  25  m.  p.  h.  cut-off  6  in., 
8  in.  and  10  in.,  the  mean  effective  pressure  was  30*5, 
51 '2  and  63*3  Ibs.  per  sq.  in.  respectively.  At 
35  m.  p.  h.  on  the  same  cut-offs,  the  m.  e.  p.  was  29'6, 
42'4,  and  48  Ibs.  per  sq.  in.,  the  boiler  pressure  being 
130  Ibs. 

A  few  typical  results  obtained  from  tests  actually 
made  on  the  road  may  now  be  given.  They  will  be 
more  conveniently  stated  in  tabular  form. 

A  series  of  coal  consumption  observations  made 
on  the  London  and  North  Western  Railway  between 
shallow  fire-box  engines  of  the  Experiment  class  and 
deep  fire-box  engines  of  the  Precursor  class,  both 
classes  working  the  heaviest  and  fastest  trains  be- 
tween Euston  and  Crewe  under  identical  conditions, 
gave  the  following  results  : 

The  Precursor  engine  ran  34,348  miles,  and  burnt 
882  tons  5  cwts.  of  coal — equivalent  to  an  average 
consumption  of  57'53  Ibs.  per  mile. 

The  Experiment  engine  ran  34,013  miles,  burning 
793  tons  8  cwts.  of  coal,  the  equivalent  average  con- 
sumption being  52'25  per  mile. 

Speeds.  The  performance  of  a  locomotive  is 
generally  associated  in  the  mind  of  the  average 
traveller  with  speed,  without  regard  to  any  other 


IX] 


PERFORMANCE  AKD  SPEEDS 


147 


. 

C    co        T 


.s    -a  d 


^ 


I  sSd  §1  Is  i 


8    45 : 

06      -* 


i  i 


1:5 

O         ^ 


CO  W5 
<N«5 

O5  tH 


II      I      IS      I 


E 


0       Q 


10—2 


148  THE  MODERN  LOCOMOTIVE         [CH. 

consideration.  There  exists  too  a  popular  vague 
idea  that  electricity  is  to  make  practicable  hitherto 
unimaginable  travelling  speeds.  This  idea  is  difficult 
to  account  for  on  any  other  basis  than  that  electricity 
is  capable  of  doing  for  us  everything  that  has  hitherto 
been  found  impossible  without  its  aid.  No  high 
travelling  speeds  have  been  attempted  commercially, 
and  the  Berlin-Zossen  high-speed  tests  in  which  a 
specially  built  racing  electric  motor  vehicle  ran  the 
distance  of  14*3  miles  in  8  minutes,  attaining  a  maxi- 
mum speed  of  210  kms.  (130  miles)  per  hour,  have 
no  smack  of  commercial  economy  about  them.  This 
experimental  car  hauled  no  load,  and  if  as  much 
money,  trouble  and  time  had  been  spent  upon  a 
high-speed  steam  racing  machine,  we  should  probably 
have  learnt  that  the  same  velocity  is  possible  with 
steam  locomotion.  The  only  condition  favourable  to 
high-speed  and  long-distance  electric  traction  is  to 
run  single  or  two  coach  flying  expresses  at  small 
time  intervals,  involving  clearing  the  line  of  all  other 
traffic.  Such  a  condition  for  the  generality  of  our 
main  lines  and  having  regard  to  the  standard  of 
comfort,  such  as  dining,  sleeping  and  heavy  baggage 
accommodation,  required  by  passengers  to-day,  puts 
the  method  out  of  court  in  favour  of  heavy  express 
trains  in  the  haulage  of  which  the  steam  locomotive 
has  the  advantage.  The  commercial  aspect  of  this 
question  is  much  too  strong  a  factor  in  railway 


ix]          PERFORMANCE  AND  SPEEDS          149 

administration  ever  to  be  sacrificed  to  any  considera- 
tion which  would  involve  the  most  serious  of  outlays 
and  the  most  doubtful  of  returns. 

The  attainment  of  high  speed  is  by  no  means  con- 
fined to  electrically  propelled  vehicles.  The  highest 
speed  ever  authentically  recorded  in  favour  of  the 
steam  locomotive  is  given  as  having  been  reached  by 
the  high-speed  engine  constructed  by  the  Prussian 
State  Railway  Department  and  exhibited  at  the 
St  Louis  Exhibition.  In  the  course  of  its  trials,  this 
locomotive  maintained  a  speed  of  82  miles  an  hour 
with  a  six-car  train,  representing  a  tonnage  of  240 ; 
a  speed  of  87  miles  an  hour  with  five  cars,  200  tons ; 
and  a  speed  of  92  miles  with  three  cars,  120  tons. 

In  this  country  it  is  doubtful  if  any  higher  speed 
has  been  authenticalty  recorded  under  modern  con- 
ditions, i.e.  hauling  a  passenger  train  of  average 
weight,  than  that  reached  on  the  Gt  Western  Railway 
by  the  No.  1  Ocean  up  special  express  on  August  30, 
1909.  The  occasion  was  the  opening  of  the  new 
harbour  at  Fishguard,  and  the  Curiard  steamship 
Mauretania  having  beaten  her  previous  best  time 
from  New  York  in  a  passage  of  4  days  4  hours 
27  minutes,  it  was  probable  that  the  Gt  Western 
Railway  would  also  try  for  a  record.  The  writer 
was  on  the  train  and  recorded  a  maximum  speed 
of  90  miles  per  hour,  which  was  verified  by  the 
representative  of  Engineering,  who  was  also  a 


150  THE  MODERN  LOCOMOTIVE         [CH. 

passenger.  A  portion  of  the  run  of  this  train  is 
plotted  on  the  accompanying  chart  (Fig.  36)  against 
the  gradient  profile.  The  engine  taking  the  train 
from  Cardiff  to  Paddington  was  the  King  Edward, 
the  first  of  a  batch  of  four-cylinder  six-coupled  non- 
compound  locomotives  now  hauling  the  fastest  trains 
of  the  company.  The  load  consisted  of  10  eight- 
wheeled  bogie  carriages  aggregating  274  tons,  and 
the  running  time  between  Cardiif  and  Paddington — 
a  distance  of  145J  miles — was  141J  minutes,  giving 
an  average  speed  of  61*6  miles  per  hour.  This  average 
speed  over  a  long  distance,  although  magnificent,  was 
surpassed  in  the  race  to  Aberdeen,  but  it  is  doubtful 
if  the  maximum,  90  miles  per  hour,  was  exceeded  on 
that  occasion.  This  occurred  in  running  down  the 
1  in  300  bank  between  Badminton  and  Wootton 
Bassett,  although  incidentally  an  equally  meritorious 
performance  was  the  23  \  miles  up  hill  from  Severn 
Tunnel  Junction  to  Badminton,  with  length  of  1  in  100, 
1  in  68,  1  in  90  up  and  ten  miles  of  1  in  300  up  in 
29  min.  55  sees,  or  at  the  rate  of  47  miles  per  hour. 

The  highest  average  speed  of  a  regular  train  was 
attained  in  the  famous  race  to  Aberdeen  in  August, 
1895.  This  contest  arose  between  the  East  Coast 
partnership,  viz.  the  London  and  North  Western  and 
Caledonian  Railways  and  the  East  Coast  Companies, 
the  Gt  Northern,  North  Eastern  and  North  British, 
and  gave  rise  to  some  very  fine  running.  The  distance 


IX] 


PERFORMANCE  AND  SPEEDS 


151 


% 


fr 


i 


m 


w 


,jl     I  >aADMINm 


WOOTTON 

easserr 


WIHDON 


152 


THE   MODERN  LOCOMOTIVE 


[CH. 


from .  Euston  to  Aberdeen  is  539f  miles  and  from 
King's  Cross  to  Aberdeen  523£ — nearly  17  miles 
less.  In  the  first  case  there  are  the  severe  climbs 
up  Shap  Fell  and  the  Beattock  Summit  to  be 
reckoned  with,  and  on  the  East  Coast  the  slacks 
necessitated  by  the  Forth  and  Tay  Bridges.  The 
following  were  the  results: — 


Date 

West  Coast 

East  Coast 

1895 

Depart 
Euston 

Arrive 
Aberdeen 

Depart 
King's  Cross 

Arrive 
Aberdeen 

Aug.  19 
,,      20 

>.      21 
.,      22 

8  P.M. 

H 

)  ) 
j> 

5-15  A.M. 
4-58    „ 
4-45    ., 
4-32    ,, 

8  P.M. 

Kace  aba 

5-31  A.M. 
5-11    „ 
4-38    ,, 
ndoned 

The  decisive  victory  gained  by  the  West  Coast 
partnership  on  the  22nd  involved  running  at  an 
average  speed  for  the  whole  distance  of  63*3  miles 
per  hour  including  stops,  or  64*1  without.  The  length 
between  Crewe  and  Carlisle  was  run  at  an  average 
speed  of  no  less  than  65*1  miles  per  hour.  The  load, 
it  is  true,  was  a  light  one  consisting  of  3  bogie  coaches 
totalling  70  tons,  but  with  the  type  of  engine  used  it 
would  manifestly  have  been  impossible  to  have  ac- 
complished such  an  achievement  with  two  or  three 
hundred  tons  behind  the  tender.  The  engines  em- 
ployed during  the  race  by  the  West  Coast  were  the 


IX] 


PERFORMANCE  AND  SPEEDS 


153 


three-cylinder  7  ft.  compounds  Coptic  and  Adriatic 
from  Euston  to  Crewe  and  thence  to  Carlisle,  the  older 
type  Precedent  class,  Hardwicke  and  Queen.  The 
Caledonian  Co.  used  their  four-coupled,  6  ft  6  in.  type. 
The  following  are  the  details  of  the  famous  run  of 
the  West  Coast  on  the  last  day  of  the  race : 


Time 

Miles 

Engine 

Average 
Speed 

Euston        dep.    8  P.M. 
Crewe          arr.     10.28 
dep.    10.30 
Preston       pass    11.16 

158 

Webb,  7  ft. 
3-cylinder 
compound 
Hardwicke 

60 

Carlisle       arr.     12.36 

141| 

Precedent  class 

65-1 

dep.    12.38| 

Perth           arr.       3.7^ 
dep.      3.9| 

i«i 

Caledonian,  6  ft.  6  in. 
4-coupled,  No.  90 

60-9 

Aberdeen    arr.       4.30 

6  ft.  6  in. 

(Ticket  Platform 
Station)       ...     4.32 

190 

4-coupled 
No.  17 

65 

It  may  here  be  recalled  that  on  Sept.  8th  of  the 
same  year,  the  three-cylinder  compound  Ionic  ran 
from  Euston  to  Carlisle  without  a  stop — 299J  miles 
in  5  hours  53  minutes,  i.e.  at  an  average  speed  of 
51  miles  per  hour,  with  a  load  of  150  tons  14  cwts. 

The  race  to  Edinburgh  in  1888  between  the  same 
companies  caused  some  sensation,  and  at  the  time 
the  performances  were  unequalled  in  railway  history. 


154 


THE  MODERN  LOCOMOTIVE 


[CH. 


A  feature  of  the  contest  was  that  the  West  Coast 
train  was  run  as  far  as  Crewe  with  old  single  engines 
of  the  Lady  of  the  Lake  class  (see  page  8)  and 
beyond  with  4-coupled  Precedents.  The  Caledonian 
also  used  a  7  ft.  single  engine,  Mr  Drummond's  famous 
No.  123.  Mr  Stirling's  superb  singles  were  used  by 
the  Gt  Northern  Co.  as  far  as  York.  The  best 
single  performance  by  the  West  Coast  train  was 
on  August  14th,  when  the  400  miles  were  covered 
in  427  minutes  of  running  time,  or  at  the  rate  of 
56J  miles  per  hour  throughout ;  that  by  the  East 
Coast  train  was  on  August  31st,  when  the  392f  miles 
were  covered  in  412  minutes,  a  speed  of  more  than 
57  miles  per  hour.  The  best  single  performances  by 
each  route  are  given  by  Mr  Ac  worth  as  follows : — 


Date 

Section 

Distance 
(miles) 

Time 
(minutes) 

Remarks 

Aug.   13 

Euston  —  Crewe 

158£ 

166 

»        7 

Crewe—  Preston 

51 

50 

— 

„        7 

Preston  —  Carlisle 

90 

90 

Over  Shap 

summit 

„       9 

Carlisle  —  E  dinburgh 

lOOf 

103 

Over  the 

Beattock 

„      25 

King's  Cross  —  Grantham 

105| 

106 

— 

.,      24 

Grantham  —  York 

82| 

88 

— 

„      29 

York—  Newcastle 

80£ 

81 

— 

.,      14 

Newcastle  —  Edinburgh 

1241 

125 

Two  engines 

on.  Best  time 

with  one,  130 

minutes, 

Aug.  31 

ix]  PERFORMANCE  AND  SPEEDS          155 

E  very-day  speeds  in  1911  were  of  a  very  high 
average  order.  A  recent  inquiry  conducted  by  the 
International  Railway  Congress  Association  brought 
to  light  the  following  facts,  as  to  railways  which  were 
in  the  habit  of  using  speeds  of  100  km.  (62  miles) 
in  regular  service. 

In  Belgium  the  speed  may  attain  110  kilometres 
(68*4  miles)  per  hour  in  cases  of  delay,  on  the  flat 
and  on  down  gradients  of  5  to  6  per  mil.,  and  on 
curves  having  a  diameter  of  at  least  900  metres 
(45  chains).  The  Baden  State  Railway  uses  speeds 
attaining  110  kilometres  (68*4  miles)  per  hour  on 
down  gradients  not  steeper  than  4'5  per  mil,  and  on 
curves  having  a  radius  of  not  less  than  1100  metres 
(55  chains) ;  in  Germany,  this  radius  is  legally  pre- 
scribed as  the  minimum  for  a  speed  of  110  kilometres 
(68*4  miles)  per  hour.  On  several  French  railways, 
speeds  of  110  to  115  kilometres  (68*4  to  71*5  miles) 
per  hour  are  attained  in  cases  of  delay,  on  down 
gradients  of  not  more  than  5  per  mil.,  and  on  curves 
having  a  radius  of  at  least  700  metres  (35  chains); 
the  .railways  in  question  are  the  Midi,  the  Paris- 
Lyons-Mediterranean  and  the  French  State.  Of  the 
railways  regularly  using  speeds  of  over  100  km. 
(62  miles)  per  hour,  France  is  represented  by  speeds 
which  may  attain  120  kilometres  (74*5  miles)  per 
hour.  This  is  the  maximum  speed  fixed  by  the 
supreme  authorities.  It  is  attained  on  up  gradients 


156  THE  MODERN  LOCOMOTIVE          [CH. 

of  not  more  than  3'3  per  mil.,  on  down  gradients  of 
not  more  than  5  per  mil.,  and  on  curves.  With  the 
Bavarian  locomotive  speeds  of  up  to  150  kilometres 
(93  miles)  per  hour  are  stated  to  have  been  attained. 

In  connection  with  the  subject  of  speed  it  is 
noteworthy  that  the  highest  velocities  have  not 
been  obtained  with  the  largest  size  driving  wheels. 
Mr  Marshall,  in  a  paper  read  before  the  Institution 
of  Civil  Engineers,  drew  attention  to  a  run  in  which 
11  successive  miles  were  covered  by  a  train  drawn 
by  an  engine  with  driving  wheels  of  5ft.  8  in. 
diameter.  If  driving  wheels  of  7  ft.  6  in.  diameter, 
with  proportionally  larger  cylinders,  could  be  worked 
at  the  same  number  of  reciprocations  per  second  (13), 
a  speed  of  112  miles  per  hour  would  be  obtained. 
Mr  Marshall  suggested  that  the  cause  of  the  practical 
limitation  in  the  speed  of  locomotives  is  not,  as 
has  been  generally  assumed,  the  steam  not  being 
able  to  escape  quickly  enough  from  the  cylinders, 
but  should  be  looked  for  in  a  slipping  of  the  driving 
wheels,  arising  from  the  effective  adhesion  weight 
being  seriously  reduced,  when  running  at  high  speeds, 
by  the  vertical  action  of  the  disturbing  forces  of 
the  balancing  masses  in  the  rotation  of  the  wheels. 
It  has  been  found  that  at  a  speed  of  74  miles  per 
hour,  a  slip  of  as  much  as  19  per  cent,  takes  place  in 
a  four-coupled  engine. 

The  idea  put  forward  that  the  main  requisites 


x]  COMPOUNDING  157 

for  obtaining  high  speed  are  an  increase  in  the 
number  of  coupled  wheels  and  a  corresponding  in- 
crease in  the  boiler  power  has  certainly  been  borne 
out  in  recent  practice,  and  is  well  evidenced  in  the 
example  given  above  of  the  run  of  the  Ocean  express 
on  the  Great  Western  Railway. 


CHAPTER  X 

COMPOUNDING 

IT  was  seen  in  a  previous  chapter  that  one  of  the 
main  causes  of  loss  in  the  working  of  an  ordinary 
engine  was  due  to  condensation  which  is  set  up  when 
the  steam  at  boiler  pressure  is  brought  into  contact 
with  the  walls  of  a  cylinder  cooled  down  by  the  low 
temperature  of  the  exhaust  steam.  The  temperature 
difference  may  be  as  much  as  140°  F.,  as  for  example 
when  the  working  pressure  is  180  Ibs.,  with  a  tempera- 
ture equivalent  of  380°  F.,  and  the  terminal  pressure 
10  Ibs.,  with  a  temperature  of  240°  F. 

High  piston  speeds  may  reduce  this  range.  Ac- 
cording to  Mr  Hughes,  when  the  piston  speed  exceeds 
600  ft.  per  minute,  the  period  is  too  short  to  render 
the  difference  of  temperature  due  to  the  interchange 
of  heat  noticeable  in  simple  and  compound  working. 


158  THE  MODERN  LOCOMOTIVE         [OH. 

One  remedy  applied  is,  as  has  been  seen,  to  super- 
heat the  steam  before  its  arrival  at  the  cylinder  ; 
another  and  older  method  is  compounding.  It  is 
not  however  with  this  sole  object  in  view  that  com- 
pounding is  resorted  to  ;  its  adoption  means  also  an 
efficiency  obtained  by  expanding  the  steam  through 
more  stages  than  is  possible  in  a  single  cylinder. 

It  can  be  shewn  that  the  efficiency  of  a  perfect 
heat  engine  may  be  measured  by  the  ratio 


where  T!  is  the  absolute  maximum  temperature  of 
the  working  fluid  in  the  engine  and  r2  the  absolute 
maximum  temperature  of  the  working  fluid  in  the 
engine.  The  cycle  cannot  be  realised  in  practice 
but  it  indicates  theoretically  and  practice  confirms, 
that  the  greater  the  difference  between  rl  and  r2  the 
greater  will  be  the  efficiency.  The  largest  difference 
is  obtained  by  compounding.  The  steam  is  admitted 
into  one  cylinder  called  the  '  high-pressure  '  cylinder, 
and  expansion  allowed  to  commence  therein,  and 
afterwards  it  is  exhausted  into  a  second  or  'low- 
pressure  '  cylinder,  where  the  expansion  is  continued. 

The  number  of  times  the  steam  is  expanded  is 
called  the  ratio  of  expansion  :  e.g.  if  it  is  expanded 
twice,  the  ratio  of  expansion  is  2  to  1. 

It  is  a  convenience  in  connection  with  the  calcu- 
lation of  horse-power  and  mean  effective  pressure 


x]  COMPOUNDING  159 

to  consider  the  total  expansion  as  referred  to  the  low- 
pressure  cylinder.  It  is  immaterial,  for  this  purpose, 
what  the  ratio  of  expansion  may  be  in  each  cylinder;  it 
is  as  though  the  whole  range  takes  place  in  the  low- 
pressure  cylinder.  If  the  capacity  of  the  latter  is, 
say  9  cubic  feet,  and  steam  is  cut  off  after  one  cubic 
foot  has  been  admitted  to  the  high-pressure  cylinder, 
then  the  ratio  of  expansion  is  9  to  1.  It  is  of  import- 
ance that  the  work  done  in  each  cylinder  should 
be  theoretically  approximately  equal.  The  exact 
ratio  of  the  volume  of  the  high-  and  low-pressure 
cylinders  is,  however,  a  somewhat  debateable  point, 
as  it  ranges  from  1  to  1*69  up  to  1  to  3  and  more. 

Owing  to  the  range  of  temperature  and  the  great 
differences  of  piston  effort  at  the  beginning  and  end 
of  the  stroke  the  practical  limit  of  expansion  in 
a  single  cylinder  is  about  three  times,  whereas,  by 
further  utilizing  the  steam  in  a  second  cylinder, 
instead  of  exhausting  it  to  the  atmosphere,  practically 
double  the  amount  can  be  obtained,  the  limitation 
being  the  requirement  of  a  certain  amount  of  pressure 
in  the  exhaust  steam  to  serve  for  the  blast. 

In  engines  of  the  stationary  and  marine  type 
expansion  is  continued  until  the  exhaust  pressure 
becomes,  by  the  employment  of  a  condenser,  that  due 
to  a  vacuum,  but  the  limitations  of  the  locomotive 
prohibit  the  use  of  such  an  apparatus.  Nevertheless 
good  results  are  obtained  in  stationary  engine  practice 


160  THE  MODERN  LOCOMOTIVE         [CH. 

with  non-condensing  compound  engines,  hence  it  was 
thought  that  the  method  had  only  to  be  applied  to 
locomotives  to  secure  its  general  adoption.  This  may 
be  said  to  be  the  case  in  the  country  where  it  was 
first  applied,  viz.  France,  and  more  or  less  generally 
on  the  continent. 

It  was  first  applied  by  Mallet  in  1878  on  the 
Bayonne  and  Biarritz  railway,  with  one  small  high- 
pressure  cylinder  and  one  large  low-pressure  cylinder. 
Mallet's  method  was  tried  with  slight  improvements 
in  1880  by  von  Borries  on  the  Hanoverian  State 
railways,  but  the  first  to  put  compound  locomotives 
into  regular  use  was  the  late  Mr  Francis  Webb  of  the 
London  and  North  Western  Railway.  In  his  earliest 
engine  two  outside  high-pressure  cylinders  were  used, 
14  in.  diameter  by  24  in.  stroke,  which  drove  outside 
cranks  on  the  trailing  wheels ;  and  one  large  low- 
pressure  cylinder,  30  in.  in  diameter  and  24  in.  stroke, 
placed  inside  the  frames  and  below  the  smoke-box, 
driving  on  to  a  crank  in  the  axle  of  the  leading  pair 
of  driving  wheels.  Thus  only  one,  and  that  a  very 
large  cylinder,  exhausted  to  the  chimney  which  gave 
these  engines  a  characteristic  'beat.'  A  large  number 
of  them  were  built  and  ran  apparently  successfully 
for  a  number  of  years,  but  on  Mr  Webb's  retirement 
they  were  gradually  withdrawn.  What  economy  was 
obtained  with  them  remains  a  secret,  but  that  they 
were  not  good  at  starting  was  obvious  to  all  observant 


x]  COMPOUNDING  161 

travellers.  The  high-pressure  cylinders  were  of 
small  diameter  and,  acting  alone,  insufficient  to 
start  a  heavy  train.  Until,  however,  they  did  start 
working,  the  low-pressure  cylinder  could  get  no 
steam.  What  usually  happened  was  that  the  wheels 
driven  by  the  high-pressure  cylinder  started  slipping 
badly,  giving  the  large  low-pressure  cylinder  more 
than  enough  steam,  so  that  this  started  working  with 
a  violent  series  of  jerks  which  continued  during 
acceleration,  and  communicated  themselves  to  the 
train,  rendering  it  very  uncomfortable  for  the  pas- 
sengers. 

Mr  Webb  afterwards  adopted  the  four-cylinder 
arrangement  on  his  compound  engines  of  the  Black 
Prince  type.  These  had  two  outside  high-pressure 
cylinders,  15  ins.  diameter,  and  two  inside  low-pressure 
cylinders,  16|  ins.  in  diameter,  both  pairs  having  a 
common  stroke  of  24  in.  They  were  all  situated  in 
line  under  the  smoke-box,  and  drove  on  to  the  first 
pair  of  coupled  wheels  and  its  axle.  The  wheels  were 
7  ft.  1  in.  diameter.  A  number  of  these  engines  are 
still  in  service. 

Mr  Webb's  example  was  followed  in  1885  by 
Mr  James  Worsdell  on  the  Gt  Eastern  Railway,  who 
used  one  high-pressure  and  one  low-pressure  cylinder 
located  between  the  frames.  His  successor,  Mr  Holden, 
however,  converted  them  all  to  the  ordinary  non- 
compound  type.  On  the  North  Eastern  Railway 

A.  L.  11 


162  THE  MODERN  LOCOMOTIVE          [CH. 

Mr  Worsdell  re-introduced  the  compound  system, 
which  his  brother,  Mr  Wilson  Worsdell,  continued, 
using  the  Worsdell-von  Borries  arrangement  of  two 
cylinders.  Steam  after  exhausting  from  the  high- 
pressure  cylinder  was  passed  round  the  smoke-box 
to  the  low-pressure  valve  chest.  A  device  known 
as  an  intercepting  valve  was  introduced  on  these 
engines,  by  which  steam  could  be  admitted  to  the 
low-pressure  cylinder  at  will  by  the  driver  when  this 
was  necessary  for  starting  purposes.  Both  the  start- 
ing and  intercepting  valves  were  operated  by  steam 
and  controlled  by  one  handle.  If  the  engine  did  not 
start  when  the  regulator  was  opened,  which  occurred 
when  the  engine  was  '  blinded,'  the  driver  pulled  the 
additional  small  handle  which  closed  the  passage 
from  the  receiver*  to  the  low-pressure  cylinder,  and 
also  admitted  a  small  amount  of  steam  to  the  low- 
pressure  steam  chest,  so  that  the  two  cylinders 
together  developed  additional  starting  power.  After 
one  or  two  strokes  of  the  engine,  the  exhaust  steam 
from  the  high-pressure  cylinder  automatically  forced 
the  two  valves  back  to  their  normal  position,  and  the 
engine  proceeded,  working  compound. 

With  the  possible  exception  of  the  Midland  Rail- 

*  A  receptacle  used  when  the  cranks  are  set  at  90°.  When  the 
h.p.  cylinder  is  exhausting,  the  port  of  the  h.p.  cylinder  has  not  yet 
been  opened  for  steam.  The  h.p.  exhaust  is  therefore  passed  into  the 
receiver  from  which  the  h.p.  cylinder  draws  its  supply. 


x]  COMPOUNDING  163 

way,  the  subsequent  history  of  compounding  in  Great 
Britain  is  limited  to  a  series  of  trial  engines  on  the 
Gt  Western,  Lancashire  and  Yorkshire,  Gt  Central 
and  Gt  Northern  railways.  On  the  Midland,  a 
number  of  three-cylinder  engines  constructed  on  the 
Smith  system  are  at  work.  Of  the  three  cylinders, 
one  is  high-pressure  and  two  low-pressure,  the  high- 
pressure  cylinder  being  placed  between  the  frames 
and  the  two  low-pressure  outside.  The  high-pressure 
cylinder  takes  steam  direct  from  the  boiler,  at  a 
pressure  of  220  Ibs.  per  square  inch,  and  this  steam 
exhausts  into  the  chest  common  to  the  low-pressure 
cylinders.  The  steam  regulator  operates  an  ingenious 
arrangement  consisting  of  a  main  and  jockey  valves. 
When  moved  over  to  start  the  engine,  high-pressure 
steam  is  admitted  simultaneously  to  the  main  steam 
pipe  and  to  the  low-pressure  auxiliary  pipe.  When 
the  main  valve  is  on  the  point  of  moving,  the  area 
of  the  passage  by  which  boiler  steam  can  pass  to  the 
low-pressure  cylinder  is  maximum,  and  further  move- 
ment of  the  handle  causes  the  main  valve  gradually  to 
close  this  opening  and  also  to  increase  the  opening 
for  the  passage  of  steam  from  the  boiler  into  the 
high-pressure  steam  pipe.  The  admission  of  boiler 
steam  to  the  low-pressure  cylinder  is  entirely  cut 
oif  by  moving  the  handle  about  30°  from  the  shut 
position,  when  the  engine  of  course  commences  to 
work  entirely  as  a  compound. 

11—2 


164  THE   MODERN  LOCOMOTIVE         [OH. 

On  the  Continent  and  in  America,  the  compound 
system  has  been  applied  to  the  locomotive  by  the 
following  methods: 

Two  Cylinders,  one  high-pressure  and  one  low- 
pressure,  driving  cranks  set  at  90°.  In  normal  working 
one  cylinder  is  supplied  with  steam  at  boiler  pressure, 
and  an  apparatus,  sometimes  automatically  operated, 
is  provided  for  admitting  boiler  steam  direct  to  the 
low-pressure  cylinder  at  starting  (Mallet,  von  Borries, 
Golsdorf,  etc.). 

Three  Cylinders.  This  system  has  found  little 
application  abroad. 

Four  Cylinders.  Two  high-  and  two  low-pressure 
cylinders  are  disposed  in  tandem  in  the  Woolff  type, 
and  operate  two  cranks  set  at  90°;  two  valve  gears 
are  employed,  the  valves  in  each  group  being  operated 
by  the  same  link.  The  addition  of  a  starting  ap- 
paratus is  not  indispensable,  but  generally  some 
provision  is  made  for  the  direct  admission  of  boiler 
steam  to  the  low-pressure  cylinder.  In  the  Vauclain 
(America)  system,  the  two  high-  and  two  low-pressure 
cylinders  are  vertically  superposed  and  drive  to  a 
common  crosshead.  Two  valve  gears  are  employed, 
and  one  valve  distributes  steam  to  each  pair  of 
cylinders. 

In  the  Adriatic  type  also,  one  valve  serves  each 
pair  of  cylinders.  The  arrangement  is  very  ingenious, 
the  cylinders  being  grouped  as  follows  : 


x]  COMPOUNDING  165 

The  two  H.P.  cylinders  are  placed  on  one  side  of 
the  engine,  one  inside  and  one  outside,  operating  two 
cranks  set  at  180°;  on  the  other  side  of  the  engine 
are  the  two  L.P.  cylinders,  one  inside  and  one  outside, 
also  driving  cranks  set  at  180°  to  each  other  and  at 
90°  to  the  H.P.  pair.  The  valve  is  placed  above  the 
outside  cylinder,  and  distributes  steam  to  its  pair  of 
cylinders  by  cross  passages. 

The  two  high-  and  two  low-pressure  cylinders  may 
drive  respectively  cranks  set  at  90°,  the  pairs  being 
at  180°  to  each  other.  The  H.P.  and  L.P.  cranks  are 
sometimes  on  the  same  axle,  as  in  the  Maffei  and 
von  Borries  systems,  or  drive  on  two  separate  coupled 
axles,  as  in  the  famous  French  type  known  as  the 
de  Glehn  and  du  Bousquet  system.  This  dispenses 
in  principle  with  the  use  of  a  starting  apparatus, 
since  the  arrangement  comprises  two  high-pressure 
cylinders  with  cranks  at  90°.  In  practice,  however, 
it  is  often  applied  owing  to  the  small  dimensions  of 
the  H.P.  cylinders  which,  in  certain  positions  of  the 
cranks,  may  be  unable  to  start  the  train.  This  is 
the  system  which  has  been  applied  to  practically 
all  the  French  engines  constructed  since  1896  and 
with  which  such  excellent  results  have  been  obtained, 
particularly  on  the  Northern  of  France  Railway. 

A  considerable  fuel  economy,  amounting  in  some 
cases  to  as  much  as  20  per  cent.,  is  definitely  admitted 
to  have  been  found  in  the  working  of  compound 


166  THE   MODERN  LOCOMOTIVE          [CH. 

engines,  and  as  M.  Sauvage  has  pointed  out,  it  is 
rather  cinder-estimating  the  merits  of  the  compounds 
to  say  that  by  their  use  the  weight  of  trains  is  in- 
creased by  one-third  with  the  same  cost  of  fuel 
over  what  is  used  with  the  best  simple  engines 
used  before.  If  not  weight,  increased  speed  is 
obtained,  and  in  many  cases  both  weight  and  speed. 
His  presentation,  which  represents  the  French  view, 
of  the  case  of  the  compound  versus  simple  fairly 
considers  all  the  circumstances.  He  states  'the 
initial  cost  of  the  compounds  is  higher,  the  expenses 
for  repairs  somewhat  greater,  but  the  increase  of 
traffic  is  such  that  the  economy  is  obvious.  A 
complete  solution  of  the  problem  would  require  a 
proof  that  the  same  results  might  not  be  obtained 
in  some  other  way.  Available  data  are  not  sufficient 
to  give  such  a  proof  in  an  incontestable  manner ; 
still,  it  seems  difficult  to  build  an  ordinary  locomotive 
quite  equal  in  every  respect  to  the  latest  compounds. 
It  is  clear  that  simple  two-cylinder  engines  might  be 
made  with  the  same  large  boiler,  and  work  with  the 
same  high  pressure,  but  it  is  nearly  as  clear  that, 
with  the  ordinary  valve  gear  of  the  locomotive, 
steam  at  such  a  high  pressure  cannot  be  utilized 
as  well  as  by  compounding ;  there  is  little  doubt 
that  the  simple  locomotive  would  require  more 
steam  for  the  same  work.  In  addition,  there  is  a 
real  difficulty  in  making  all  the  parts  of  the  simple 


x]  COMPOUNDING  167 

engine  strong  enough  to  stand  without  undue  wear 
the  greatest  stresses  resulting  from  the  increased 
pressure  on  large  pistons,  although  this  difficulty 
may  be  overcome.  An  opinion  which  seems  to  prevail 
is  that  compound  locomotives  may  be  economical 
during  long  runs,  but  that  their  advantage  is  lost 
when  they  stop  and  start  frequently,  owing  to  the 
direct  admission  of  steam  to  the  low-pressure 
cylinders  at  starting.  This  opinion  is  rather  too 
dogmatic,  and  the  question  requires  some  con- 
sideration. In  many  cases,  with  four-cylinder  com- 
pounds, the  tractive  power  necessary  for  starting 
from  rest  is  obtained  without  this  direct  admission, 
and  steam  is  admitted  in  that  way  only  for  the  first 
revolution  of  wheels.  The  engine  is  then  worked 
compound,  but  in  full  gear  for  all  cylinders.  Of 
course,  steam  is  not  so  well  utilized  as  with  a  proper 
degree  of  expansion  in  each  cylinder,  but,  even  in 
that  case,  the  compound  compares  favourably  with 
a  simple  locomotive  working  in  full  gear.'  Under 
these  circumstances  it  is  difficult  to  account  for  the 
unpopularity  of  compounding  with  British  locomotive 
engineers.  Meagre  as  they  are,  the  results  published 
in  this  country  incontestably  shew  the  compound  to 
be  a  more  efficient  and  economical  machine  than  the 
simple  engine.  Against  this  there  is  to  be  placed  the 
statement  made  in  some  quarters,  that  the  additional 
cost  of  maintenance,  due  to  increased  complication, 


168  THE   MODERN  LOCOMOTIVE         [OH. 

more  than  neutralizes  the  advantages  gained  in  fuel 
saving.  It  is  difficult  to  see,  however,  that  the 
adoption  of  three  or  four  cylinders  working  non- 
compound  does  not  introduce  the  same  increase  in 
cost  of  maintenance  ;  which  may  also  be  said  of  the 
addition  of  superheating  apparatus  to  engines  of  the 
ordinary  two-cylinder  type  involving  the  installation 
of  mechanical  lubricators,  piston  valves  and  numerous 
accessories.  M.  Demoulin,  of  the  Western  Railway 
of  France,  has  even  stated  that  the  capital  involved 
is  more  than  that  required  for  compounding,  for  the 
same  number  of  cylinders  ;  and  it  will  probably  be 
admitted  that  superheating  when  applied  under  the 
most  favourable  conditions,  although  yielding  an 
economy  analogous  to  that  which  results  from  com- 
pounding, is  nevertheless  inferior  thereto. 

This  does  not  mean  that  continental  engineers 
are  neglecting  superheating ;  on  the  other  hand  they 
are  using  it  in  combination  with  compounding,  which 
seems  to  be  highly  advantageous  if  it  can  be  obtained 
without  adding  greatly  to  the  complexity  of  the 
whole  machine.  It  is  difficult  to  reconcile  conflicting 
opinions,  hence  the  writer  has  limited  himself  to  a 
simple  statement  of  the  question. 

The  Future.  Turbine  locomotives  have  been  ex- 
perimented with  in  Germany  and  by  Mr  Reid,  of  the 
North  British  Locomotive  Co.,  but  have  not  met  with 
much  success.  The  efficiency  of  the  turbine  depends 


x]  COMPOUNDING  169 

essentially  upon  adequate  condensing  arrangements, 
for  which  air  cooling  or  the  limited  quantity  of 
cooling  water  capable  of  being  carried  on  an  engine 
is  quite  inadequate.  The  future  may  see  an  increased 
application  of  the  water  pick-up  system,  which  would 
considerably  advance  matters  in  this  direction,  as 
not  only  would  the  necessary  partial  vacuum  then 
be  maintained,  but  the  condensed  steam  could  be 
pumped  back  into  the  boiler  at  a  high  temperature. 
The  provision  of  such  facilities  would  equally  favour 
compounding,  and  enable  results  to  be  obtained 
therefrom  comparable  with  those  realised  in  marine 
and  stationary  practice.  On  the  other  hand  no 
exhaust  steam  would  be  available  for  the  purpose 
of  creating  a  draught. 


BIBLIOGRAPHY 


LAKE.     The  Locomotive  Simply  Explained.     Percival  Marshall  & 

Co.     A  useful  introduction  to  the  subject. 
HUGHES.     The  Construction  of  the  Modern  Locomotive.     E.  and 

F.  N.  Spon.     A   practical   work   dealing  with  methods  of 

manufacture  and  erection. 
PETTIGREW.      A    Manual    of    Locomotive    Engineering.      Chas. 

Griffin  &  Co.    The  standard  English  work. 
ANON.     The  Locomotive  of  To-day.     The  Locomotive  Publishing 

Co.    An  excellent  work  dealing  very  fully  with  constructional 

details. 

DEMOULIN.     1.  Traite  de  la  Machine  Locomotive.     2.  La  Loco- 
motive Actuelle.     Beranger,  Paris.     A  standard  work. 
NADAL.     Locomotives   a  Vapeur.     Octave  Doin,    Paris.      Very 

valuable    for    its    clear    mathematical  treatment   of  many 

problems  untouched  in  other  treatises. 
GARBE.    Die  Dampflokomotiven  der  Gegenwart.    Julius  Springer, 

Berlin.     The  standard  German  treatise. 
PENDRED.     The  Railway  Locomotive.     Constable  &  Co.     A  very 

attractive  work,   containing  much  useful    information    not 

found  elsewhere.     Treatment  not  too  technical. 
DALBY.     The  Balancing  of  Engines.     Edwin  Arnold  &  Co. 
VON    BORRIES.     Die    Lokomotiven    der    Gegenwart.      Kreidel, 

Wiesbaden. 


BIBLIOGRAPHY  171 


PAPERS   AND  ARTICLES. 

ASPINALL.  Train  Resistance.  Proceedings  Institution  Civil 
Engineers.  1901-2. 

MARSHALL.  The  Evolution  of  the  Locomotive  Engine.  Proceed- 
ings Institution  Civil  Engineers.  1897-8. 

SAUVAGE.  Recent  Locomotive  Practice  in  France.  Proceedings 
Institution  Mechanical  Engineers.  1900. 

CHURCHWARD.  Large  Locomotive  Boilers.  Proceedings  Institu- 
tion Mechanical  Engineers.  1906. 

HUGHES.  Locomotives  designed  and  built  at  Horwich  with  some 
Results.  Proceedings  Institution  Mechanical  Engineers. 
1909. 

-  Compounding  and  Superheating  in  Horwich  Locomotives. 
Proceedings  Institution  Mechanical  Engineers.     1910. 

SAMS.      Modern    Locomotive    Construction.      The    Engineering 

Review.     1908. 
SUMNER.     The  Power  of  a  Locomotive  Boiler.    The  Engineering 

Review.     1910. 

-  Coal  Consumption  on  Locomotives.     The  Engineering  Re- 
view.    1909. 

STROUDLEY.  The  Construction  of  Locomotive  Engines.  Pro- 
ceedings Institution  Civil  Engineers.  1885. 


INDEX 


Accident,  Salisbury,  136 ;  Double- 

bois,  133 
Acworth,  154 
Adhesion,  86 
Ashpan,  16 

Atlantic  type  engine,  54,  124 
Axle  boxes,  117;  radial,  118 
Axles,  116 

Balancing,  127 
Baltic  type,  10,  53,  54 
Bissel,  119,  140 
Blast,  25,  27;  variable,  28 
Bogie,  119,  121,  140 
Boiler,  12, 18;  Brotan,  49  ;  Cone, 
42;  Stayless,  48;  Schneider,  50 

Carlisle,  non-stop  run  to,  153 
Castings,  steel,  114 
Cataract  gear,  106 
Centrifugal  couple,    126;    force, 

125,  135 

Charles  Dickens,  4 
Coal  consumption,  40,  146 
Combustion,  30 
Compounding,  11,  157,  165 
Compounds,  three -cylinder,  160, 

163,   164;    four-cylinder,    161, 

164 

Conductivity,  rate  of,  38 
Connecting  rods,  117 
Cooke,  C.  J.  Bowen,  4,  73 
Coiiard,  M.,  134 


Coupling  rods,  117 
Cylinders,  10,  93,  95 

Dalby,  128 
De  Glehn,  165 
Demoulin,  168 
Derailments,  135 
Dimensions,  increase  in,  6,  8 
Drawbar  pull  at  different  speeds, 

85 

Drummond,  D.,  5,  53,  76 
Du  Bousquet,  165 
Dynamometer  car,  142 

Electric  traction,  2 

Feed- water  heating,  76 

Firearch,  16 

Firebars,  15 

Fire-box,     14;      Belpaire,     43; 

Jacobs-Schupert,   47 ;    Marine 

type,    52  ;     Wagon    top,    46 ; 

Wootten,  44 
Foundation  ring,  17 
Frames,    112;     bar,    113;     box 

girder,    115;    combined    plate 

and  bar,  115 
Fuel,  30 ;  heating  values  of,  33 

Gauge,  limit  imposed  by,  6 
Grate  area,  10 
Gravity,  centre  of,  135 
Gt  Bear,  9 


INDEX 


173 


Hammer  blow,  130 

Heat  losses,  37 ;  engine,  efficiency 

of,  158 

Heating  surface,  10 
Heaton,  127 
Holden,  59,  161 
Hughes,    72,    78,   98,    131,    157, 

161 

Injectors,  exhaust  steam,  60 
Ivatt,  84 

Jenny  Lind,  7,  8 
Johnson,  S.  W.,  5 

Lag  effect,  141 

Lagging,  29 

Lap,  102 

Latent  heat,  64 

Lead,  102 

Locomotion,  the,  8,  9 

Logarithms,  hyperbolic,  93 

McConnell,  127 
Maffei,  165 
Mallet,  160 
Marshall,  156 

Non-stop  runs,  9 
North  British    Locomotive   Co., 
115,  168 

Oil  burning  apparatus,  57 
Oscillation,  133 

Performance  locomotive,  141 

Pettigrew,  80 

Piston  rod,  94 

Pistons,  96 

Pony  truck,  119 

Ports,  94 

Power  absorbed  by  locomotive,  86 


Precedent  class,  L.N.W.R.,  4, 7,  8 
Pressure,  mean  effective,  90 
Purdue  University,  144 

Eace  to  Aberdeen,  150;  to  Edin- 
burgh, 153 

Radiant  heat,  38 

Railways  :  —  Birmingham  and 
Gloucester,  127 ;  Caledonian, 
28,  153  ;  Dutch,  72  ;  Glasgow 
and  Southwestern,  106;  Great 
Central,  11;  Great  Eastern, 
59,  106,  161 ;  Great  Northern, 
78,  84, 141 ;  Great  Western,  10, 
124,  133,  144,  149 ;  Lancashire 
and  Yorkshire,  10,  72,  78,  98, 
107,  131;  London  and  North 
Western,  73,  86,  107,  118,  160; 
London  and  South  Western, 
10,  53,  54,  55,  76,  137;  Mid- 
land, 11,  163;  North  British, 
54;  North  Eastern,  142,  162; 
Northern  of  France,  52,  54, 
80,  165;  Pennsylvania,  144 

Receiver,  162 

Reid,  168 

Resistance,  79 ;  due  to  curves, 
82;  due  to  gradient,  81; 
formulas  for,  80 ;  internal,  84 

Reversing  gear,  power,  106 

Rocket,  the,  4,  13 

Sauvage,  M.,  166 

Sensible  heat,  63 

Sisterson,  86 

Slide  valve,  96 

Smoke  box,  23,  24,  26 

Spark  arrester,  28 

Speeds,     146;     in     1911,     155; 

Ocean  express,  G.W.R.,  149 
Stability,  125 
Stay  bolts,  17,  18 


174 


INDEX 


Steam,  saturated,  64,  65;  total 
heat  of,  66;  wet,  67;  chests, 
95 

Stephenson,  3 

Stirling,  James,  106 

Stirling,  P.,  5,  8,  9,  141,  154 

Stoking,  35 

Stroudley,  W.,  5 

Super-elevation,  139 

Superheated  steam,  total  heat  of, 
68 

Superheating  apparatus,  73 

Testing  plant,  144 

Tests,  table  of,  147;  Berlin- 
Zossen,  148;  Purdue  Uni- 
versity, 71 

Thermal  storage,  77 

Tractive  effort,  87 

Trevithick,  3 

Tubes,  13,  29  ;  Serve,  21 


Tube-plate,  16 
Turbine  locomotive,  168 

Valve     gear,     101;     Joy,     107; 

Stephenson,   104;  Walschaert, 

109 
Valves:   Lentz,  99;   piston,  97; 

rotary,  101;  poppet,  99;  slide, 

96 ;  Stumpf,  100 
Von  Borries,  160 

Water  softening,  56;  tubes,  48,  55; 
Brotan  system,  49 ;  Riegel,  52 

Webb,  F.,  118,  141,  160 

Whale,  86 

Wheel  arrangement,  121;  dia- 
gram, 123 

Wheels,  115 

Wilson,  Carus,  83 

Worsdell,  J.,  161 

Worsdell,  W.,  162 


CAMBRIDGE  :    PRINTED    BY    JOHN    CLAY,    M.A.    AT    THE    UNIVERSITY    PRESS. 


THIS  BOOK  IS  DUE  ON  THE  LAST  DATE 
STAMPED  BELOW 


AN  INITIAL  FINE  OF  25  CENTS 

WILL  BE  ASSESSED  FOR  FAILURE  TO  RETURN 
THIS  BOOK  ON  THE  DATE  DUE.  THE  PENALTY 
WILL  INCREASE  TO  SO  CENTS  ON  THE  FOURTH 
DAY  AND  TO  $1.OO  ON  THE  SEVENTH  DAY 
OVERDUE. 


• 


31  1938 


REC'D  LD 


*   ^-SPM 


wjrrfr 


tr* 


DEC  19  '65 


JUL    7 191:-; 


'  U       I U